Tesis Cristina Medina

Sensorarrays
forEnology
appli
cat
ions:
Usingnanos
cience
f
orgrapeanalys
is
Cr
ist
inaMedinaPlaz
a
Doct
oralThes
is
ESCUELA DE INGENIERÍAS INDUSTRIALES
DEPARTAMENTO DE QUÍMICA FÍSICA Y QUÍMICA

INORGÁNICA
TESIS DOCTORAL:
Sensor arrays for Enology applications:

Using nanoscience for grape analysis
Presentada por Cristina Medina Plaza para optar al

grado de Doctora por la Universidad de Valladolid
Dirigida por:
Prof. Dra. María Luz Rodríguez Méndez

Valladolid, 2016
AUTORIZACIÓN DE LA DIRECTORA DE TESIS
Dª María Luz Rodríguez Méndez, con D.N.I: nº 50305983-S,

Catedrática de Universidad del Departamento de Química Física
y Química Inorgánica de la Escuela de Ingenierías Industriales,
como Directora de la Tesis Doctoral titulada “Sensor arrays for
Enology applications: Using nanoscience for grape analysis”,
presentada por Dª. Cristina Medina Plaza, alumna del programa
“Doctorado en Física” impartido por la Facultad de Ciencias,
autoriza la presentación de la misma, considerando que cumple
todos los requisitos para ello.
Valladolid, 15 de Abril de 2016
La Directora de la Tesis
Fdo. María Luz Rodríguez Méndez

TABLE OF CONTENTS
List of Abbreviations and Acronyms……………………………………………………1-4
Justifications and Objectives…………………………………………………………5-10
CHAPTER 1-Introduction..……………………………………………………………11-44
1.1 GRAPES……………………………………………………………………………..14
1.1.1 Chemical composition of the grapes………………….…………..15
1.1.2 Berry ripening……………………………………………………..………….16
1.2 ELECTRONIC TONGUE………………………………………………………..19
1.2.1- Mass sensors……………………………………………………..………….20
1.2.2- Optical sensors……………………………………………..……………….21
1.2.3-Electrochemical sensors………………………………..……………….23
1.2.4-Chemical modifiers…………………………….………………..…………26
1.2.5-Preparation of voltammetric sensors..……………………………32
References………………………………………………………………………………36
CHAPTER 2- Analysis of Organic Acids and Phenols of Interest in the

Wine Industry using Langmuir–Blodgett Films Based on Functionalized
Nanoparticles……………………………………………………………….……..………45-68
2.1 INTRODUCTION………………………………………………………………...48
2.2 MATERIALS AND METHODS……………………………………………….49
2.2.1-Chemicals…………………………….…………………………………………49
2.2.2-Equipment…………………………………………….……………………….50
2.2.3-Preparation of nanoparticles………………….………………………50
2.2.4-Preparation of LB films…………………………….…………………….50
2.2.5- Electrochemical studies…………………………………………………51
2.3 RESULTS AND DISCUSSION…………………………………………………51
2.3.1-Structural characterization……………………………………………..51
2.3.2-Electrochemical response in simple electrolytes…………….52
2.3.3-Electrocatalytic effect of SDODAuNP-LB towards organic

acids…………………………………………………………………………….………….53
2.3.4-Electrocatalytic effect of SDODAuNP-LB towards phenolic

acids……………………………………………………………………………….……….58
2.3.5-Response of SDODAuNP-LB towards mixtures of organic

acids and phenols………………………………………………………..………....61
2.3.6-Electrode repeatability and reproducibility…………..………..62
2.4 CONCLUSIONS……………………………………………………………………62
References………………………………………………………………………..…….64
CHAPTER 3- Structural and Electrochemical Properties of Lutetium Bis-

Octachloro-Phthalocyaninate Nanostructured Films. Application as
Voltammetric Sensors…………………………………………………………….……69-94
3.1 INTRODUCTION…………………………………………………………………71
3.2 MATERIALS AND METHODS……………………………………………….73
3.2.1-Chemicals……….……………………………………..……………………….73
3.2.2-Langmuir, LB and LS films…………………..….……………………….73
3.2.3-Film characterization………………………………...…………….…….74
3.3 RESULTS AND DISCUSSION…………………………………………………75
3.3.1-Langmuir films……………….………………………….…………………..75
3.3.2-LB and LS films. Growth monitored by UV-Vis

absorption……………………………………………………………………………….77
3.3.3-LB and LS films. Molecular organization determined by

FTIR…………………………………………………………………………………………78
3.3.4-Surface Enhancement Resonance Scattering

(SERRS)……………………………………………………………………………………80
3.3.5-Electrochemical properties………….…………………………..…….82
3.3.6-Electrochemical detection of catechol……………………………83
3.3.7-Dynamic behavior…………………………………….…………….……..87
3.3.8-Limits of detection……………………………….………………….….…88
3.4 CONCLUSIONS……………………………………………………………………89
References………………………………………………………………………………91
CHAPTER 4- Synergistic Electrocatalytic Effect of Nanostructured Mixed

Films Formed by Functionalized Gold Nanoparticles and
Bisphthalocyanines………………………………..…………..……………………..95-136
4.1 INTRODUCTION……………………………………………..………………….97
4.2 MATERIALS AND METHODS……………………………………..………..99
4.2.1-Langmuir monolayers and Langmuir-Blodgett films…..…..99
4.2.2-Films characterization…………………………………………………..101
4.3 RESULTS AND DISCUSSION………………………………………………102
4.3.1-Langmuir monolayers…………………………………………………..102
4.3.2-Langmuir-Blodgett films characterization…………………….104
4.3.3-Electrochemical characterization………………………………….107
4.3.4-Electrocatalytic properties. Voltammetric sensors……….109
4.4 CONCLUSIONS…………………………………………………………………112
References…………………………………………………………………………….115
CHAPTER 5- Bioelectronic Tongue Based on Lipidic Nanostructured

Layers Containing Phenol Oxidases and Lutetium Bisphthalocyanine for
the Analysis of Grapes………………………………………………….………….119-142
5.1 INTRODUCTION……………………………………………………………….122
5.2 MATERIALS AND METHODS……………………………………………..124
5.2.1-Chemicals…………………………………………..…………………………124
5.2.2-Langmuir and Langmuir-Blodgett films

characterization…………………………………………………………………….124
5.2.3-Electrochemical measurements………………………..………….125
5.2.4-Phenols and grapes……………………..…………………………..…..125
5.2.5-Statistical analysis. Data treatment………………..…………….126
5.3.1-Langmuir monolayers and Langmuir-Blodgett

films………………………………..…………………………………………………….126
5.3.2-Electrochemical response towards phenols………………....129
5.3.3-Array of sensors. Discrimination of phenolic

compounds……………………………………………….…………………………..135
5.3.4-Array of sensors. Response towards grapes………............136

5.4 CONCLUSIONS…………………………………………………..…………….137
References…………………………………………………………………………….139
CHAPTER 6-Array of Biosensors for Discrimination of Grapes According

to Grape Variety, Vintage and Ripeness…………………………………..143-162
6.1 INTRODUCTION……………………………………………………………….145
6.2.1-Chemicals……………………………………………………………………..147
6.2.2-Grape samples……………………………………………………………..148
6.2.3-Langmuir-Blodgett films……………………………………………….148
6.2.4-Electrochemical measurements……………………………………149
6.2.5-Statistical analysis…………………………………………………………149
6.3.1-Characterization of Langmuir-Blodgett films…………………150
6.3.2-Discrimination of must obtained from different varieties of

grapes harvested in two consecutive vintages………………………152
6.3.3-Monitoring the grape ripeness by means of the

bioelectronic tongue……………………………………………………………..155
6.3.4-Multiparametric correlations between the bioelectronics

tongue and the classical chemical analysis…………………………….156
6.4. CONCLUSIONS………………………………………………………………..157
References…………………………………………………………………………….159
CHAPTER 7- Electrochemical Characterization of Dilithium
Phthalocyanine Carbonaceous Electrodes………………………………..163-180
7.1 INTRODUCTION……………………………………………………………….165
7.2.1-Chemicals……………………………………………………………………..167
7.2.2-Equipment……………………………………………………………………168
7.3.1-Electrochemical characterization………………………………….168
7.3.2-Analysis of simple electrolytes……………………………………..172
7.3.3-Analysis of citric acid…………………………………………………….174
7.4 CONCLUSIONS…………………………………………………………………176
References…………………………………………………………………………….177
CHAPTER 8-Electronic Tongue Formed by Sensors and Biosensors

Containing Phthalocyanines as Electron Mediators. Application to the
Analysis of Red Grapes…………………………………………………………….181-206
8.1 INTRODUCTION……………………………………………………………….183
8.2.1-Grape samples……………………………………………………………..185
8.2.2-Reagents and solutions…………………………………………………186
8.2.3-Preparation of the electrodes……………………………………….186
8.3.1-Response of the array of CPE sensors modified with

phthalocyanines……………….…………………………………………………..189
8.3.2-Response of the array of biosensors formed by CPEs
modified with phthalocyanines and tyrosinase………….………....192
8.3.3-Response of the array of biosensors formed by CPEs

modified with phthalocyanines and glucose oxidase…….….…..194
8.3.4-Response towards musts………………………………………………196
8.3.5-Data treatment………………………………….…………………..…….198
8.4 CONCLUSIONS……………………………………….……………………..…201
References…………………………………………………………………………….202
CHAPTER 9- The Advantages of Disposable Screen-Printed Biosensors

in a Bioelectronic Tongue for the Analysis of Grapes………….…...207-230
9.1 INTRODUCTION…………..…………………………………………………..209
9.2 MATERIALS AND METHODS……………..…………………….….……212
9.2.1-Chemicals……………………………………..…………………...………..212
9.2.2-Grape samples…………………………………..……………...…………212
9.2.3-Screen-printed electrodes and biosensors………..………….213
9.2.4-Electrochemical measurements………………………..………….215
9.2.5-Statistical analysis…………………………………………………..……215
9.3 RESULTS AND DISCUSSION …………………………………………….216
9.3.1-Measurements in glucose and catechol………………..………216
9.3.2-Measurements in musts. Optimization of the experimental

conditions……………………………………………………………………………..218
9.3.3-Bioelectronic tongue. Statistical analysis…………………..….223
9.4 CONCLUSIONS…………………………………………………………..…….226
References…………………………………………………………………………….228
CHAPTER 10- Nanoscale Au-In Alloy-Oxide Core-Shell Particle as

Electrocatalysts for Efficient Hydroquinone Detection……………231-252
10.1 INTRODUCTION………………………………………..……………………233
10.2 MATERIALS AND METHODS……………………………………………235
10.2.1-Synthesis of AuxIn1-x alloy nanoparticles………………..……235
10.2.2-Characterization of Au-In alloy nanoparticles…………..…235
10.2.3-Electrochemical characterization…………………………....…236
10.3 RESULTS AND DISCUSSION……………………………….……………236
10.3.1-Structural characterization………………………………..……….236
10.3.2-Electrocatalytical properties……………………….….………….241
10.3.3-Dynamic behavior………………………………….…….…………….247
10.3.4-Limits of detection………………………….…………….……………248
10.4 CONCLUSIONS…………………………………………………….…………250
References……………………………………………………………….……………251
CHAPTER 11- Final Conclusions……………………………………………….253-258
Resumen en español………………………………………………………………259-276
1- INTRODUCCIÓN…………………………………………………………………261
2- OBJETIVOS…………………………………………………………………………264
3- METODOLOGÍA Y RESULTADOS…………………………………………265
3.1-Sensores nanoestructurados mediante la técnica de

Langmuir-Blodgett…………………………………………………………………265
3.2-Sensores de pasta de carbono…………………………………………270
3.3-Sensores serigrafiados…………………………………………………….272
3.4-Sensores modificados con nanopartículas core-shell……….273
4- CONCLUSIONES…………………………………………………………………275
Anex……………………………………………………………………………………………….279
List of Abbreviations and Acronyms
AA Arachidic Acid
AFM Atomic Force Microscopy
AuNP Gold Nanoparticle
BAM Brewster Angle Microscopy
C Cabernet-Sauvignon
Caf Caffeic Acid
Cat Catechol
CNT Carbon Nanotubes
CoPc Cobalt Phthalocyanine
CP Conductive Polymer
CPE Carbon Paste Electrode
CuPc Copper Phthalocyanine
CV Cyclic Voltammetry
DC Direct Current
DOD Dodecanethiol
DODAB Dimethyldiocatdecylammonium bromide
DP Diffraction Pattern
EDP Electrodeposition
EDS Energy Dispersive X-ray Spectroscopy
EF Enhancement factor
1
Enz Enzyme
ET Electronic Tongue
FDH D-Fructose Dehydrogenase
FTIR Fourier Transform Infrared
G Garnacha
GCE- Glassy Carbon Electrode
GOx Glucose Oxidase
GPH Graphene
HPLC High Performance Liquid Comatography
HQ Hydroquinone
ISS Low-energy Ion-Scattering Spectroscopy
ITO Indium Tin Oxide
Lac Laccase
LB Langmuir-Blodgett
LbL Layer by Layer
Li2Pc Dilithium phthalocyanine
LnPc2 Lantanoid Bisphthalocyanine
LOD Limit of Detection
LS Langmuir-Schaeffer
LSPR Localized Surface Plasmon Resonance
LuPc2 Lutetium Bisphthalocyanine
2
LuPc2Cl32 Bis[2,3,9,10,16,17,23,24-
octachlorophtalocianate] lutetium (III)
LV Latent Variables
M Mencía
MIP Molecular Imprinted Polymer
MONP Metal Oxide Nanoparticle
MPc2 Metallophthalocyanine
MWCNT Multiwall Carbon Nanotubes
NiONP Nickel Oxide Nanoparticle
NPc Naphthalocyanine
PB Prussian Blue
PBS Buffer Phosphate
Pc Phthalocyanine
PC Principal Component
PCA Principal Component Analysis
PLS Partial Least Squares
PP Prieto Picudo
Pyr Pyrogallol
QCM Quartz Crystal Microbalance
RMSEC Root Mean Square Error of Calibration
RMSEP Root Mean Square Error of Prediction
RRS Resonance Raman-Scattering
3
RSD Relative Standard Deviation
SAM Self-Assembly Monolayer
SAuNP (11-mercaptoundecyl)tetra(ethylenglycol)
functionalized gold nanoparticle
SAW Surface Acoustic Wave
SDODAuNPs n-dodecanethiol functionalized gold

nanoparticle
SEM Scanning Electron Microscopy
SERRS Surface-Enhanced Resonance Raman Scattering
SPE Screen Printed Electrode
T Tempranillo
TEM Transmission Electron Microscopy
TPI Total Polyphenol Index
Tyr Tyrosinase
UHV Ultrahigh Vacuum
UV-Visible Ultravioleta Visible
Van Vanillic Acid
XPS X-ray Photoelectron Spectroscopy
ZnPc Zinc Phthalocyanine
4
JUSTIFICATION&OBJECTIVES
5
6
Justification and Objectives
The present thesis entitled “Sensor arrays for Enology applications:

Using nanoscience for grape analysis” has been developed in the Group
of Sensors of the University of Valladolid (UVaSens). This document is
presented to obtain the International Ph.D. Mention. For this reason and
according to the regulations of the University of Valladolid it is written in
English and includes a summary of the results and the general
conclusions written in Spanish.
The Group of Sensors works simultaneously on basic and applied

research. The basic research is focused on the preparation and
characterization of new electrochemical sensors and their application
for complex liquids analysis such as wines, oils, beers… The applied
research is devoted to the development of electronic tongues where
arrays of sensors coupled to the appropriate pattern recognition
software are used to the organoleptic characterization of food and
beverages. In the last years, the group has focused its research on the
enology sector carrying out analysis of musts and wines. These researchs
are developed with the collaboration of Instituto Tecnológico Agrario de
Castilla y León (ITACyL), Bodega Cooperativa de Cigales and Estación
Enológica de Castilla y León.
The success of previous research on the enology area made the group to
go in depth in this field. This thesis tackles the development of new
sensors based on different electroactive materials by means of different
techniques. Different strategies have been foreseen to improve the
performance and selectivity of the sensors by including enzymes and
new electrocatalytic materials. Efforts have been undertaken to develop
new nanostructured and biomimetic biosensors that can improve the
performance of the system. These new sensors have been combined to
form arrays of sensors and biosensors applied for the analysis of grapes
quality and ripening control.
According to these general ideas, the objectives of this thesis are:
1- To desig new voltammetric sensors and biosensors using

different electrocatalytic materials (nanoparticles, phthalocyanines) and
7
enzymes (tyrosinase, laccase, glucose oxidase, D-fructose

dehydrogenase).
2- To develop new methodologies to deposit the sensing materials

and to immobilize the enzymes. Methods ranging from simple
techniques which allow preparing simple and cheap sensors to more
sophisticated techniques which produce nanostructured sensors with
better performances but at higher cost will be employed.
3- To test these new sensors towards analytes present in must such

as antioxidants, sugars and organic acids.
4- To combine sensors and biosensors in order to get different

(bio)electronic tongues able to discriminate grapes with different
organoleptic characteristics.
5- To develop a bioelectronic tongue containing enzymes devoted

to the analysis of phenols and sugars applied to monitor the ripening
process.
The work presented here is placed on the cross-road between enology

and material science and has been financed by MINECO (AGL2009-
12660/ALI and AGL2012-33535), Junta de Castilla y Leon (VA-032U13)
and by a scholarship (PIF-UVa).
The present memory is organized as follows:
Chapter 1 presents a general overview about electronic tongues

to help the understanding of the basis of this dissertation. This chapter
presents a short synopsis about the composition of grapes. A section
focused on electronic tongues and the different techniques employed to
develop them is also included.
Chapters 2 to 10 encompass the experimental block of the

dissertation. Each chapter is self-consistent and presents 4 sections:
“introduction”, “materials and methods”, “results and discussion” and
“conclusions”.
8
Chapters 2 to 6 are centered on the development of

nanostructured sensors prepared using Langmuir-Blodgett (LB)
technique. Different strategies to prepare good quality LB films from
different materials (gold nanoparticles, phthalocyanines, enzymes) are
presented. Chapter 2 research was developed in the collaboration with
Prof. Kawai from Tokyo University of Science (Japan) and is focused on
the preparation of LB films and nanostructured sensors containing
functionalized gold nanoparticles. Chapter 3 is devoted to the
development of sensors containing a bisphthalocyanine by means of LB
and LS. A methodology to co-immbolize two electrocatalytic materials in
the LB film and the evaluation of their synergistic effect are presented in
chaper 4. The results presented in this chapter were obtained on
collaboration with researchers from UNESP (Brazil). Similarly, a method
to co-immobilize an enzyme with the electron mediator and the effect of
the biomimetic environment is discussed on chapters 5 and 6. In all the
cases, structural and electrochemical characterization of the films and
sensors is presented. Arrays of sensors combined with pattern
recognition software are presented in chapters 5 and 6. These
bioelectronic tongues are dedicated not only to the analysis of standard
solutions but also to the analysis of grapes.
Despite the excellent performance of nanostructure (bio)sensors,

classical techniques were used to obtain cheaper sensors to form arrays
that could be easily applied in the industrial environment. This idea has
been developed in chapters 7 to 9.
Chapters 7 and 8 are focused on Carbon Paste Electrodes.

Different carbonaceous materials are tested on chapter 7 whereas
chapter 8 is devoted to the evaluation of different phthalocyanines as
electron mediator in a bioelectronic tongue. The latest is used to the
discrimination of musts.
Chapter 9 presents a bioelectronic tongue made by disposable

screen printed electrodes devoted to the analysis of phenols and sugars.
Discrimination capability towards red grapes juice is also showed.
9
During my 3-month research stay at Brookhaven National Laboratory

(NY,USA) I joined Dr. Eli Sutter’s group. This research group is focused on
the development of new nanomaterials with environmental or energy
applications. The research done during those 3 months is presented in
chapter 10.
The synthesis of Au-In core-shell nanoparticles and their

structural and electrochemical characterization is showed in chapter 10.
Their electrocatalytical properties towards the analysis of hydroquinone,
a hazardous byproduct of some industries, are discussed and compared
with classical nanoparticles.
Chapter 11 presents the final conclusions of the work.
10
CHAPTER 1
Introduction
Introduction
Agrofood sector constitutes one of the most important bases of the

economy in Castilla y León, manufacturing almost a fourth of the
industrial production of the region. In the national ranking, Castilla y
León is on the top positions of food and agriculture production.
In the last years, enology sector has been confirmed as one of the most
dynamic parts of the agrofood sector in Castilla y León. This sector has
experienced a huge expansion due to the strict selection of the seeds,
improvement of the cultivation, severe control of the berries ripening,
winemaking process under firm quality controls and settled-down
commercialization.
The region presents more than 75000 hectares of vineyards with more
than 5750 workers. It presents 530 cellars which produce around 200
million of liters of wine per year. The region has ten designation of origin
(denominación de origen D.O.) and the 86% of the vineyards are
dedicated to this sign of quality. In 2013, the economic value of this
sector reached more 738 million of euros being 134 generated by
exportation1.
The main objectives of the technologies carried out during winemaking

process are focused on an effective extraction of phenolic compounds
which are responsible of the color, taste and aroma of the wines and a
control of sugar fermentation which is directly related to the alcoholic
degree of the final product.
As raw material of the wine industry, quality of grapes must be

controlled since it directly influences the quality of the final product. In
order to obtain grapes with the best characteristics, not only the ripe
berry need to be controlled but also the ripening process. A good
understanding of grape composition and ripening is essential to
comprehend the process of winemaking and making better quality wine.
13
Chapter 1: Grapes
Due to the research line of this thesis is devoted to the analysis of grapes
and musts the following section describes the structure of the grapes
and the classical methods used to analyze their composition.
1.1 GRAPES
The structure of the grape berry can be divided into three parts: skin,
flesh, and seed2 (Figure 1.1). The chemical composition of each of these
parts is different and directly influences the final quality of the grapes
and therefore the wine3.
Under the outer layer of skin cells, grapes present a flattened layer of
cells that tend to accumulate phenolic compounds in relatively high
concentrations as the grape berry matures4. These compounds strongly
influence aroma, flavor and color attribute. Hence, they play a key role
in determining final juice and wine quality and sensory characteristics.
The compounds concentrated in these cells include tannins, which

strongly influence mouthfeel such as astringency, flavor such as
bitterness and many specific taste notes. Most of these compounds are
accumulated during early berry development, prior to veraison. The
concentration of tannins does not appear to be much altered in the skin
during the later stages of grape berry maturation, but the chemical
structure of the tannins is modified. These changes lead to less
astringency, less bitterness and a smoother mouthfeel. Pigments, mainly
anthocyanins in red grapes, start being accumulated in these cells at
veraison.
The cells that form the berry flesh contain the juice and are the primary
sites for the accumulation of sugars during grape berry ripening. Acids
and phenolic compounds are also concentrated in these cells, with the
primary acids being tartaric and malic acid. In contrast with those
located in the skin, acids and tannins located in the flesh tend to
decrease in per-berry concentration during berry ripening.
The seed coat also contains relatively high concentrations of tannins.

Similar to the tannins and phenols found in the flesh, these tannins are
14
Chapter 1: Grapes
reduced in concentration as the berries mature5. In addition, phenols

responsible for bitterness become less soluble/extractable.
Figure 1.1. Structure of a grape berry partially sectioned to show internal parts.
Principal compounds found in each part are also pointed.
1.1.1-Chemical composition of the grapes

Grape juice consists of 70 to 80% water and many dissolved solids. These
soluble solids include numerous organic and inorganic compounds. The
important groups of compounds, from the winemaking point of view,
include sugars, organic acids and phenolic compounds.
A large portion of the soluble solid in grapes is sugars. Glucose and

fructose are the main sugars in the juice6 and their concentration varies
between 150 to 250 g/L. In unripe berries, glucose is the predominant
sugar. At the ripening stage, glucose and fructose are usually present in
equal amounts (1:1 ratio). In overripe grapes, the concentration of
fructose exceeds that of glucose.
Glucose and fructose are converted by the yeast to alcohol and carbon
dioxide during the fermentation process and their concentration is
15
Chapter 1: Grapes
directly related to the amount of alcohol produced. Hence, sugar levels

determine the price and the quality of the grapes.
Next to sugars, organic acids are the most abundant solids present in
grape juice. They are very important components of juice and wine. As
they are responsible for the sour taste and have a marked influence on
wine stability, color and pH. The principal organic acids found in grapes
are tartaric, malic and in a small extent, citric acid7. During the early
period of berry growth, concentration of tartaric and malic acids
increases in the fruit. With the onset of ripening, while sugar
accumulates in the berry, the acid concentration decreases. Generally
the reduction in malic acid is greater and consequently mature grapes
contain more tartaric than malic acid.
Phenolic compounds are important constituents of grapes and wine.

Following sugars and acids, they are the most abundant elements
present in grapes. Phenolic compounds are a group of molecules that
are structurally diverse and that play a vital role in wine's color and
flavor. They are involved in browning reactions in grapes and wines and
they also play a key role in the aging and maturation of the final product.
The phenolic compounds are primarily located in the seeds and skins of
the berry8,9. Phenolic compounds present great interest in the food
industry due to their antioxidant capacity and their potential health
benefits. Their detection and assessment can be carried by means of
fluorescence, absorbance and high performance liquid chromatography
(HPLC) among others. Due to their redox properties they can also be
detected by electrochemical methods.
The two main molecule families included in this group of compounds are
anthocyanins and tannins. Anthocyanins are pigments and they are
responsible for the red and purple color of the grapes and wines. During
processing and aging, the tannins polymerize making them insoluble;
their precipitation leads to a decrease in wine's astringency.
16
Chapter 1: Grapes
1.1.2-Berry ripening
Three stages of grape berry development have been identified10. During
stage I, starting at fruit set, berries expand in volume and accumulate
solutes such as tartaric and malic acids, but little sugars. At stage II,
called the lag phase, berries have reached half of their final size and
continue to accumulate acids and tannins. Stage III starts at veraison,
when berries change color, accumulate sugars and metabolize acids.
After veraison, significant changes occur including decrease on malic
acid and seed tannins content and increase on glucose and fructose
concentration (Figure 1.2).
Figure 1.2. Diagram showing berries size and color at the different stages of
development.
There are some correlations that theoretically determine the optimum

ripeness of the berry called “ripeness indexes”. They consist on certain
formulation proposed to estimate the optimum maturation of the berry
and fix the harvesting day.
“Physical maturation indexes” determine quantitatively some

characteristics of the berry ripening such as color and weight of the
bunch.
17
Chapter 1: Grapes
“Chemical maturation indexes” are based on the analytical

determination of several characteristic compounds whose concentration
increase or decrease during ripening process. The traditional criteria
usually used to determine the day of harvesting are the weight of berry,
sugar content and pH.
Once the grapes have been crushed and the musts have been obtained
and clarified using the same methods that will be employed on the
winemaking process, sugar content, pH, total acidity, tartaric and malic
acid concentrations and total polyphenol index are analyzed using
traditional methods11.
In those criteria, aromatic compounds, fundamental in the winemaking

process, and polyphenolic content, an important parameter in the
elaboration of red wines are not taken into account. Nowadays, the
study of polyphenolic content is taking on a great importance in the case
of red winemaking process and some strategies to asses it have been
developed.
Regarding sensory analysis, panels of trained experts are required.

Panels need a considerable amount of resources, time and money and
suffers from some drawbacks like discrepancies due to human fatigue or
stress and clearly cannot be used for online measurements. The
“Estación Enológica de Haro” has developed a sensory method to
control grape ripeness taking into account four kinds of maturity:
sugars/acidity, flesh aroma, skin aroma and tannins12. Hence, the
implementation of analytical systems in fruit juice processing and quality
control that could improve the actual instrumentation and will permit
freshness evaluation, shelf life, and authenticity assessment and
quantitative analysis is relevant.
In recent years, a new class of instruments called electronic tongues has

been successfully applied to the analysis of wines13,14 including quality
control15-18, ageing control19,20 or detection of fraudulences21 among
others. The use of arrays of sensors is of particular interest in the field of
wines due to the importance of antioxidants, which are responsible of
18
Chapter 1: Electronic Tongue
their organoleptic characteristics and also for their health benefits.

Electronic tongues could also be used for the analysis of grapes.
1.2 ELECTRONIC TONGUE

According to the IUPAC, an electronic tongue is a multisensor system
which consists of a number of low-selective sensors and cross-sensitivity
to different species in solution and an appropriate method of pattern
recognition and/or multivariate calibration for data processing22-28.
The pattern recognition techniques consist of four sequential stages:

signal pre-processing, dimensionality reduction, prediction and
validation. The signal pre-processing prepares the feature vector for
future processing. It includes compensation for sensor drift, scaling of
the data and extracting representative parameters. The intrinsic
complexity of the signals generated by voltammetric sensor arrays can
difficult the data processing. One solution to simplify the high
dimensionality is to employ a feature extraction stage such as the kernel
method29.
A dimensionality reduction stage projects this initial feature onto a

lower dimensional space. This is usually done using a non-supervised
technique such as Principal Component Analysis (PCA). Using PCA, it is
possible to discriminate between samples with different characteristics.
The resulting low dimensional feature vector is the one used to solve a
given prediction problem, typically classification, regression or
clustering. In regression tasks, the goal is to establish a predictive model
from a set of independent variables (i.e. sensor responses) to a second
set of variables that are the properties of the sample analysed (e.g.
concentration, quality). They are usually carried out using Partial Least
Squares (PLS) regression models30.
The heart of any electronic tongue is the array of sensors. Sensors play a
decisive role in the performance of an electronic tongue, for this reason
many efforts have been dedicated to the development of new sensors
with improved characteristics. In general, sensors feature two functional
19
components: a recognition element to provide selective/specific binding

with the analyte and a transducer component for signaling the binding
act. Sensors efficiency relies on these two components for the
recognition process in terms of response time, signal/noise ratio,
selectivity and limits of detection. Therefore, designing sensors with
higher efficiency depends on the development of new materials and
preparation techniques in order to improve recognition and
transduction processes. Sensors used in electronic tongues can use
several measurement principles including mass, optical or
electrochemical transduction.
1.2.1-Mass sensors
Mass sensors are miniaturized solid-state devices which exploit the
piezoelectric effect. Most of the mass sensors developed in order to
analyze complex liquid samples are based on quartz crystal
microbalance (QCM) principles. A QCM is a mass-sensitive detector
based on an oscillating piezoelectric quartz crystal that resonates at a
fundamental frequency. The changes in the resonance frequency are
related to the changes in the mass of the substances that are adhered
onto the surface of the electrode31 (Figure 1.3). In quartz resonators the
acoustic wave can be propagated through the bulk material or/and on
the surface. Thus there are many possible modes of operation. Although
the durability of mass sensors is comparable with other sensors used in
electronic tongue systems, their selectivity is limited due to their
working principle (i.e. adsorption).
Figure 1.3. Working principle of a mass sensor.
20
The sensing materials which modify the surface of the electrodes are
indispensable for achieving highly sensitive and selective detection of a
specific chemical. For instance, polymer-coated QCM sensors have been
employed for the detection of organic compounds32, detection and
quantification of metal ions in aqueous solution has been carried out
using coatings of imprinted polymers33,34,silica mesoporous materials35
and sol gel films36 among others.
Arrays of sensors based on QCM principles have been devoted to the

analysis of different kinds of samples such as water37,38 or grapes39. A
system based on surface acoustic wave (SAW) sensors could correctly
classify basic taste substances without a selective chemical coating, but
measuring different mechanical and electrical characteristics (viscosity,
permittivity, conductivity). In addition, acoustic sensors placed in a
microfluidic cell have been used for thedetermination of pH, the
classification of beverages and the estimation of freshness and fat
content in milk40. QCM sensors coated with lipid membranes have been
applied to beer recognition41.
1.2.2- Optical sensors

Optical sensors are based on the interaction between light and matter to
determine some properties of the latest. Such sensors include an
excitation source, a sensing element, a photodetector and the electronic
circuitry. The light source is tuned to the specific analytical wavelength
to obtain the best sensitivity. The sensor can present a chemosensitive
layer, e.g. a polymeric membrane containing appropriate indicator
molecules. That indicator changes its optical properties when exposed to
a target analyte influencing the absorbance or fluorescence intensity of
the membrane.That change is monitored by the detector (usually a
photodiode or photomultiplier) whose aim is to convert the optical
signal into an electric one (Figure 1.4).
21
Figure 1.4. Working principle of an optical sensor.
There are many possible modes of operation of optical sensors including

the acquisition of fluorescence intensity42,43, absorbance44-46 or
reflectance47,48.
Optical electronic tongues involve several optical sensors, usually based

on polymeric microspheres with a chemically modified surface which
allows covalent bonding to various receptors46. The microspheres can be
placed in microwells fabricated in silicon structures with microchannels,
where the samples are injected. The microspheres modified with various
receptors are organized in a sensor array whose signals can be recorded
with the use of a CCD camera. The change of microsphere colors or
fluorescence caused by the interaction with the analyte creates a
characteristic pattern, which can be analyzed with pattern recognition
methods. Thanks to a wide variety of compounds that can be used to
modify the microsphere surfaces, lots of substances can be analyzed:
pH50,51, dissolved oxygen52, chlorine53 and metals54,55 among others.
22
1.2.3-Electrochemical sensors
Electrochemical sensors (including potentiometric, amperometric,
voltammetric and impedimetric sensors) are the most widely used
sensing units in electronic tongues because of their specificity, high
sensitivity, short response time and simple operation56,57.
1.2.3.1-Potentiometric sensors
In potentiometric sensors, the concentration of certain ions in a solution
can be quantified by measuring the membrane potential created across
an ion selective membrane at zero-current conditions. The potential
created at the membrane/solution interphase depends on the nature of
the electrode material and on the composition of the solution.
Potentiometric sensors can be prepared from different materials,
membranes and techniques27,58.
The first electronic tongue system that became commercially available

was potentiometric and it was presented by Toko and coworkers59. The
proposed sensor array –a multichannel electrode called a “taste sensor”-
consisted of potentiometric electrodes with lipid-polymeric membranes
(PVS membranes with lipid derivates). Similarly to lipid layers in the cell
walls in taste buds, synthetic lipid membranes were capable of
perception of the five basic tastes60, thanks to that feature they could
differentiate between substances exhibiting various tastes.
Potentiometric sensor arrays have been widely used to analyze complex

samples such as honey61,62, fruit juices63,64, wines68,66, dairy products67,68,
coffee69 and tea70 among others.
1.2.3.2-Impedimetric sensors
Impedance spectroscopy has also been used as a transduction method
to analyze wines. Electrodes modified with various organic materials
(including conducting polymers, perylenes, phthalocyanines or carbon
nanotubes) have demonstrated their capability to detect molecules
present in wines71. The presence of insecticides in water has also been
analyzed by means of molecular imprinted polymer (MIP) sensors72.
Rotaxanes have been used to detect toxic metals such us Hg2+, Cu2+ and
23
Pb2+ 71. A composite sensor aray based on carbon nanotubes and carbon
black dispesed in polymeric matrices and doped polythiophenes was
used to discriminate basic tastes74.
1.2.3.3-Amperometric sensors
The simplest form of amperometric detection is single-potential or
direct current (DC) amperometry. A voltage (potential) is applied
between two electrodes positioned in the column effluent. The
measured current changes when an electroactive analyte is oxidized at
the anode or reduce at the cathode. The applied potential can be adjust
to maximize the response towards the analyte of interest while
minimizing the response for interfering analytes.
Amperometric electrodes arrays have been employed to analyze wines75

and wastewater quality76 among others.
1.2.3.4-Voltammetric sensors
Voltammetric sensors are based on the same principles as
potentiometric sensors but in this case the potential is variable. The
curves obtained as a result of the analysis are called voltammograms
and the peaks shown are associated with the oxidation and reduction of
the molecules present in the solution, being their intensity proportional
to the concentration. Voltammetry or voltamperometric technique
involves different techniques of electrochemical analysis (cyclic
voltammetry, square wave voltammetry and pulse differential
voltammetry) making this kind of system more versatile.
Voltametric electronic tongues have been widely used in the enology

industry either analyzing musts76-79 or wines80,81. Other kinds of
beverages have been also analyzed by means of electronic tongue
including beer82, fruit juice, tea, milk, and coffee83.
The main success in voltammetric electronic tongues has been obtained

using chemically modified electrodes. They were developed for the first
time by our group84 and this strategy has been adopted by many
groups85-88.
24
The modification of the electrode surface with electroactive and

electrocatalytic materials (e.g. phthalocyanines, conducting polymers or
nanoparticles) provides electrodes with improved selectivity89-91.In this
case voltammograms show redox peaks associated not only with the
solution but also with the electrode (Figure 1.5). Simultaneously, the
interactions between the electrode and the solution improve
extraordinarily the selectivity of the electrodes. Such interactions
include among others: i) the oxidant or reducing character of the
solution that can modify the oxidation potential of the electrode
material; ii) the electrocatalytic activity of the electrode material that
can facilitate the oxidation of the compounds solved in the analyzed
solution; iii) the acid or basic character of the solution can
protonate/deprotonate the electrode; iv) the nature and concentration
of ions present in the solution that diffuse inside the sensing layer to
maintain the electroneutrality. These interactions produce changes on
the intensities and positions of the redox peaks and are the reason of
the important cross-selectivity achieved with these sensors. Metal
nanoparticles, polymeric materials, graphene, carbon nanotubes,
perylenes and phthalocyanines among others, have been used as
modifiers in order to get sensors with different/complementary
analytical properties.
Figure 1.5. Working principle of a modified voltammetric sensor.
25
1.2.4- Chemical modifiers

The following sections present the chemical modifiers used in of this
work.
1.2.4.1-Metal nanoparticles
Nanoparticles are defined as particles between 1 to 100 nm whose
physicochemical properties differ from the bulk counterpart due to their
small size. A reduction in the particle size causes three main changes: i)
structural changes where cell parameters could be altered even
resulting in non-stoichometric particles; ii) the electronic properties can
also be changed, due to the quantum confinement effect; iii) they
provide high surface-volume ratio, increasing the surface area for
electron exchange. All these properties can be tuned by varying their
size, shape and surrounding chemical environment.
*Gold Nanoparticles (AuNPs)- AuNPs present physical and chemical

attributes that make them excellent scaffolds for the fabrication of novel
sensors92. They can be synthesized straightforward and can be made
highly stable. The binding between the sensor and the analyzed
molecule can alter physicochemical properties of transducer AuNPs,
such as plasmon resonance absorption, conductivity, redox behavior…
that can generate detectable response signals. In addition, they present
excellent electrocatalytic properties due to enhancement on the
surface-volume ratio. The fact that AuNPs can be functionalized with a
wide range of ligands leads to the detection of small molecules or non
easily detectable targets.
*Metal Oxide Nanoparticles (MONPs)- MONPs can be used in

electroanalysis of different kinds of analytes93. Their magnetic properties
can be employed for the separation of analytes while the
semiconductive properties of certain MONPs can also be used for the
electrochemical detection of selected analytes. Many MONPs have been
used in electrochemical analysis such us nickel, iron, zinc, titanium,
cerium and manganese oxides due to their electrocatalytic behavior. The
use of mixed metal oxides has also become increasingly prevalent due to
26
the limited methods to increase the activity of an oxide containing single

metal oxide.
*Core-shell Nanoparticles- Core-shell NPs are composed of two or more

materials. This type of nanoparticles can be broadly defined as
comprising a core (inner material) and a shell (outer layer material).
These particles can consist of a wide range of different combinations in
close interaction including inorganic/inorganic, inorganic/organic,
organic/inorganic and organic/organic. The choice of the shell material
of the core-shell NP is generally dependent on the end application and
use. Nanoparticles coated with a functional material such us noble
metal, semiconductor or appropriate oxide increase the physical
properties (i.e. optical, catalytic activity, electrical, magnetic and
thermal) of the combined particles compared with the pure core
particles. Regarding electrochemistry applications mainly noble metal
cores and metal oxide shells have been used for the analysis.
1.2.4.2-Phthalocyanines
The compound known as phthalocyanine (Pc) is a symmetrical
macrocycle formed by four isoindol units (Figure 1.6).
Figure 1.6. Phthalocyanine ring.
The simplest phthalocyanine is the metal-free Pc (H2Pc).

Metallophthalocyanines (termed as MPc) consist of a phthalocyanine
ring coordinated with a metallic ion (+2) placed in the central cavity
(Figure 1.7.a). A large variety of phthalocyanine derivatives can be
obtained by modification of the aromaticity of the ring system. As an
27
example, Figure 1.7.b illustrates the structure of a family of Pc

derivatives called naphthalocyanines (NPc) characterized by extended
conjugated structures. Phthalocyanines can also form sandwich type
rare-earth complexes where a lanthanoid ion is coordinated with two
phthalocyanine rings (LnPc2) (Figure 1.7.c). Heteroleptic analogues
containing different tetrapyrrole ligands have also attracted
considerable interest (Figure 1.7.d)94. Double or triple decker complexes,
usually exhibit strong interactions resulting in intriguing electronic
and optical properties.
Figure 1.7. a) Metallophthalocyanine. b) Naphthalocyanine. c) Lanthanoid

bisphthalocyanine. d) Heteroleptic phthalocyanine-porphyrin.
Phthalocyanine structures permit chemical modification through change

of the central metal ion, through symmetric and asymmetric substitution
of the main phthalocyanine skeleton and through axial ligation to the
central metal. This wide range of possibilities explains the huge number
of phthalocyanine molecules synthesized until now and the increasing
number of Pc molecules tested as sensitive materials.
The properties which render metal phthalocyanine species valuable as

electrochemical sensing materials are related to their semiconducting
properties and electrochemical behavior95,96.
*Semicoductivity: MPcs which are p-type semiconductors when doped

with oxygen with conductivities ranging from 10-10 to 10-12 S·cm-1 at
300K. Particularly interesting is the case of LnPc2 that have particularly
high intrinsic conductivities (=10-6-10-3 S·cm-1 at T=300K). This behavior
28
is due to their neutral radical state and the strong electronic interaction
between the two tightly bound Pc rings.
*Electrochemical properties: Phthalocyanines have attracted

considerable attention for their electrochemical, electrocatalytic and
electrochromic properties. The Pc ring is an 18 electron aromatic system
that in the neutral state carries two negative charges Pc(-2). Electron
structure of phthalocyanines allows both the oxidation of the Pc ring by
one or two electrons forming Pc(-1) and Pc (0), and the reduction of the
Pc ring by one to four electrons, to yield Pc(-3), Pc(-4), Pc(-5) and Pc(-6).
In addition variation of the oxidation state of the central metal ion can
also occur. The electrochemical potentials at which the oxidations or the
reduction occur depend on the nature of the metal ion, the class of
phthalocyanine and the presence of substituents in the phthalocyanine
ring. Changes in the oxidation state are usually accompanied by changes
in the electronic absorption spectra and LnPc2 are well known as
electrochromic materials.
1.2.4.3- Enzymes
In order to make sensors more specific for the analysis of target
molecules, enzymes are deposited on the surface of the electrode,
obtaining sensor units called biosensors. Enzymes are macromolecular
biological catalysts97. They accelerate chemical reaction by decreasing
the activation energy. As the rest of catalysts, enzymes are not
consumed on the reaction process and they do not alter the equilibrium
between species. Enzymes activity can be affected by other molecules
called inhibitors that can decrease enzymes activity or denaturalize
them. Generally, enzymes are globular proteins formed by linear chains
of amino acids that folds in a three-dimensional structure. The sequence
of these amino acids specifies the structure which in turn, determines
the catalytic activity of the enzyme. The reaction takes place in a small
part of the enzyme called active site in which the substrate is hold by the
amino acids around while the reactions take places. This makes the
enzymes specific for one reaction, as the other molecules do not fit into
the active site. Enzymes’ name is derived from the substrate or the
29
chemical reaction they catalyze, with the word ending –ase, e.g.
oxireductases are the enzymes that catalyzes oxidation/reduction
reactions. In this research, two phenol oxidases (tyrosinase and laccase),
glucose oxidase and D-fructose dehydrogenase have been used in order
to improve the selectivity of the sensors.
Phenol oxidases are copper containing-enzymes that catalyze

hydroxylation of phenols producing the corresponding benzoquinones.
Tyrosinase and laccase are two of the most known phenol oxidases.
*Tyrosinase (Tyr)- Tyr carries out the oxidation of o-diphenols using

dioxygen (O2). Hydrogens removed from the phenol combine with
oxygen to form water. Tyr can be isolated from a wide variety of plants,
fungal and animal species. These different origins confer them great
diversity in terms of structural and chemical properties. No common Tyr
protein structure occurring across all species has been found (Figure
1.8).
Figure 1.8. Three dimensional structure of tyrosinase.
*Laccase (Lac)- Lac is found in many plants, fungi and bacteria species. It
presents 4 copper atoms that are bound in three different sites,
presenting diverse functions. Lac acts on p-phenols and similar
molecules, carrying out a one-electron oxidation. As oxidative enzyme it
requires oxygen as a second substrate. Copper type I is the active site
where the phenol is oxidized and type II and type III is where the oxygen
is reduced producing water. It can be polymeric, and the enzymatically
active form can be a dimer or trimer.
30
*Glucose oxidase (GOx)- GOx catalyzes the oxidation of glucose to

hydrogen peroxide and D-glucono-o-lactone which then hydrolyzes to
gluconic acid (Figure 1.9). GOx is a dimeric protein covered with
carbohydrate chains as it acts outside cells. The active site where
glucose binds is a deep pocket. Glucose presents equilibrium between α
and β anomers and at pH 7 the proportion is c.a. 37% and 63%
respectively. GOx binds specifically to β-D-Glucose and does not react
with α-D-Glucopyranose however all the glucose in solution is oxidized
due to the equilibrium is driven towards β side as it is consumed in the
reaction.
Figure 1.9. Reaction process of the oxidation of glucose by Glucose oxidase.
*D-Fructose dehydrogenase (FDH)- FDH is a membrane enzyme that

catalyzes the oxidation of D-fructose to 5-dehydro-D-fructose. It belongs
to the family of oxidoreductases and it is isolated from Gluconobacter,
an acetic acid bacter98.
1.2.5- Preparation of voltammetric sensors: Deposition

techniques
Voltammetric sensors and biosensors can be prepared using a variety of
techniques. Depending on the method used, different surface
morphologies can be obtained which have a direct effect on the sensing
behavior of the sensor. Methods range from simple techniques which
allow preparing simple and cheap sensors to more sophisticated
31
techniques which produce nanostructured sensors with better

performances but at higher cost.
1.2.5.1-Classical sensors
*Casting: In this simple method, a drop of the solution of the sensitive
material is deposited on the electrode surface. Then, the solvent
evaporates leaving behind a layer of the sensing material99,100.
*Adsorption: Adsorption of the modifier molecule on graphite or carbon

materials is one of the most popular methods due to its simplicity. The
adsorption of the molecules on the electrode surface is achieved by
immersing the electrode in a solution of the corresponding molecule in
an appropriate solvent for several minutes101.
*Carbon Paste Electrodes (CPE): CPEs are prepared by mixing graphite

(or other carbonaceous material such as carbon nanotubes) with the
modifier and a mineral oil or an epoxy resin13,102,103. This mixture is
introduced in a tube and a metallic wire is used as a contact.
*Spin-Coated films: The material contained in a solution is deposited

using centrifugal forces. The first stage is the deposition of the coating
fluid onto the substrate. Then, the substrate rotates at a constant speed.
This causes the spreading of the solution and a gradual fluid thinning
until the solvent evaporates completely (Figure 1.10). The thickness of
the film depends on the concentration and the viscosity of the solution
and on the rate of rotation of the substrate. These characteristics make
spin-coating one of the most promising methods to be used in cheap
commercial sensors104.
Figure 1.10. Spin coating process.
32
*Screen printing: Layers of the different materials such as graphite,

metal oxides or phthalocyanine derivates are printed as thick pastes by
pressing the paste through a pattern on a screen. This method has been
widely used for the preparation of electrochemical sensors. Screen
printed electrodes containing the working electrode, the reference
electrode and the counter electrode on the same device are
commercially available101,105.
1.2.5.2-Nanostructured sensors
Nanostructured thin films have shown great potential on improving the
sensitivity of chemical sensors. The first reason is that ultra-thin films
present high surface uniformity and enhanced surface to volume ratio
allowing the analyte molecules to be adsorbed or desorbed from the
molecular sites more readily. Secondly, the organization of the structure
at the nanometric level causes the film properties to differ from those
obtained with the same materials in the form of thick films. Finally, the
control of the film structure allows miming biological environment
leading to the development of biosensors with improved sensitivity and
better working. Typical methods used to fabricate nanostructured films
include self-assembly, layer by layer or the Langmuir-Blodgett
technique.
*Self-Assembled Monolayers (SAMs): The term self-assembling refers to

the spontaneous formation of self-organized monolayers adsorbed on
solid inorganic substrates. SAMs are created by chemisorptions of head
groups onto the substrate from vapor or liquid phase. The strength of
the interactions depends on the substrate and on the head groups. SAM
can be defined as a head group, tail and a functional group (Figure 1.11).
Common head groups include silanes, thiols, phosphonates, etc.106.
Figure 1.11. Self-Assembled Monolayer.
33
*Layer by Layer method (LbL): Multilayers can be prepared by depositing

alternatively cationic and anionic compounds or polyelectrolytes with
wash steps in between (Figure 1.12). This LbL self-assembly method is
easy, fast and the structure, layer composition and thickness of the films
can be accurately controlled. The LbL technique provides a tool to
construct extremely interesting structures combining a variety of
reactants such as Au nanoparticles, polyallylamine and
phthalocyanines39,107.
Figure 1.12. Layer by layer process.
*Langmuir-Blodgett (LB) films: LB films are formed by spreading a

solution of the film forming molecule onto the water surface contained
in a Langmuir trough. Upon compression, the molecules are oriented at
the interface giving rise to an ordered monolayer. This floating
monolayer can be transferred onto a solid substrate by dipping the
substrate perpendicularly to the water sub-phase. Repeated dippings
allow multilayers to be obtained where the thickness can be controlled
by the number of dipping cycles. Monolayers are usually composed of
amphiphilic molecules with hydrophilic head and hydrophobic tail. LB
technique has been widely used to prepare mixed films spreading
different kinds of molecules (phthalocyanines, lipids, nanoparticles…)
onto the subphase79, 108, 109.
It is well known that different types of sensors show different structures

and number of effective sites. For this reason, different sensors show
different performances. This aspect will be investigated in this work.
34
Chapter 1: References
References
1. Consejería de Agricultura. Junta de Castilla y León.
2. Hidalgo Togores, J. Tratado de Enología I, 2003. Ed. Mundi-Prensa.

Madrid (Spain).
3. Fanzy, C. Enología: Fundamentos científicos y tecnológicos, 2002. Ed.

Mundi-prensa. Madrid (Spain).
4. Moutounet, M., Cheynier, V., Sarni-Manchado, P. Los compuestos

fenólicos in Enología: Fundamentos científicos y tecnológicos, 2002, 114-
1136. Ed. Mundi-prensa. Madrid (Spain).
5. Kennedy, J.A., Matthews M.A., Waterhouse, A.L. Phytochem. 2000,

55, 77-85.
6. Ribereau-Gayon, P., Dubourdieu, D., Doneche, B., Lonvaud, A.

Handbook of Enology. The Microbiology of Wine and Vinifications, 2005.
Wiley, New York (USA).
7. Vincent, J.E. Let’s make wine!, 2011. lulu.com
8. Darias-Martin, J., Carrillo, M., Diaz, E., Boulton, R.B. Food Chem. 2001,
73, 217-220.
9. Lerno, L.A., Reichwage, M., Ponangi, R., Hearne, L., Block, D.E.,
Oberholster, A. Am. J. Enol. Vitic. 2015, 66, 444-453.
10. Keller, M. The Science of Grapevines – Anatomy and Physiology,

2010. Ed. Elsevier. Philadelphia (USA).
11. OIV. Compendium of International Methods of Analysis of Wines and

Musts. 2013.
12. Melendez, E., Iñiguez Crespo, M., Puras, P. Cuaderno de campo.

2011, 34-39.
35
13. Medina-Plaza, C., Revilla, G., Muñoz, R., Fernandez-Escudero, J.A.,

Barajas, E., Medrano, G., de Saja, J.A., Rodriguez-Mendez, M.L. J.
Porphyrins Phtahlocyanines. 2014, 17,1-11.
14. Moreno-Codinachs, L., Kloock, J.P., Schöning, M.J., Baldi, A., Ipatov,
A., Bratov, A., Jiménez-Jorquera, C. Analyst. 2008, 133, 1440-1448.
15. Peris, M., Escuder-Gilabert, L. Anal. Chim. Acta. 2013, 804, 29-36.
16. Buratti, S., Ballabio, D., Giovanelli, G., Dominguez, C.M., Moles, A.,
Benedetti, S., Sinelli, N. Anal. Chim. Acta. 2011, 697, 67-74.
17. Chung, S., Park, S.T., Park, S.H., Kim, J.Y., Par, S., Son, D., Bae, Y.M.,
Cho, S.I. Sensors. 2015, 15, 18197-18208.
18. Costa, A.M.S., Sobral, M.M.C., Delgadillo, I., Cerdeira, A.,

Rudnitskaya, A. Sensor Actuat. B. 2015, 207, 1095-1103.
19. Apetrei, C., Apetrei, I., Nevares, I., del Alamo, M., Parra, V.,
Rodriguez-Mendez, M.L., de Saja, J.A. Electrochim. Acta. 2007, 52, 2588-
2594 .
20. Gay, M., Apetrei, C., Nevares, I., del Alamo, M., Zurro, J., de Saja, J.A.,
Rodriguez-Mendez, M.L. Electrochim. Acta. 2010, 55, 6782-6788.
21. Parra, V., Arrieta, A., Fernandez-Escudero, J.A., Rodriguez-Mendez,

M.L., de Saja, J.A. Sensors Actuat. B. 2006, 118, 448-453.
22. Vlasov, Y., Legin, A., Rudnitskaya, A., Di Natale, C., D'Amico, A. Pure
Appl. Chem. 2005, 77, 1965-1983.
23. Winquist, F. Microchim. Acta. 2008, 163, 3-10.
24. Rodriguez-Mendez, M.L., de Saja, J.A., Medina-Plaza, C., Garcia-

Hernandez, C. Electronic tongues for the organoleptic characterization of
wines” in Electronic tongues and noses in food science. 2016. Ed.
Elsevier. Philadelphia (USA).
25. Escuder-Gilabert, L., Peris, M. Anal. Chim. Acta. 2010, 665, 15-25.
36
26. Riul, A. Jr., Dantas, C.A.R., Miyazaki, C.M., Oliveira, O.N.Jr. Analyst.
2010, 135, 2481-2495.
27. Ciosek, P., Wroblewski, W. Sensors. 2011, 11, 4688-4701.
28. Del Valle, M. Sensors. 2011, 11, 10180-10186.
29. Parra, V., Hernando, T., Rodríguez-Méndez, M.L., de Saja, J.A.

Electrochim. Acta. 2004, 49, 5177-5185.
30. Kirsanov, D., Mednova, O., Vietoris, V., Kilmartin, P.A., Legin, A.
Talanta. 2012, 90, 109–116.
31. Sauerbrey, G. Z., Physik, 1959, 206-212.
32. Fan, X., Du Sens, B. Sensors Actuat. B. 2011, 160, 724-729.
33. Tokuyama, H., Kitamura, E., Seida, Yoshimi. Talanta. 2016, 146, 507-
509.
34. Yang, Z.P., Zhang, C.J. Sensors Actuat. B. 2009, 142, 210-215.
35. Lou, H., Zhang, Y., Xiang, Q., Xu, J., Li, H., Xu, P., Li, X. Sensors Actuat.
B. 2012, 166-167, 246-252.
36. Su, P.G., Hung, F.C., Lin, P.H. Mater. Chem. Phys. 2012, 130-136.
37. Erbahar, D.D., Gurol, I., Gumus, G., Musluoglu, E., Ozturk, Z.Z.,
Ahsen, V., Harbeck, M., Sensors Actuat. B. 2012, 562-568.
38. White, C., Pejcic, B., Myers, M., Qi, X. Sensors Actuat. B. 2014, 70-77.
39. Garcia-Hernandez, C., Medina-Plaza, C., Garcia-Cabezon, C., Martin-

Pedrosa, F., del Valle, I., de Saja, J.A., Rodriguez-Mendez, M.L. Sensors.
2015, 15, 29233-29249.
40. Cole, M., Sehra, G.S, Gardner, J.W. Varadan, V.K. Proceedings of IEEE
Sensors. 2002, 1, 237-241.
37
41. Kaneda, H., Takashio, M., Okahata, Y. J. Am. Soc. Brew.Chem. 2005,
63, 89-94.
42. Aziz, A.A.A., Seda, S.H., Mohammed, S.F. Sensors Actuat. B. 2016,
223, 566-575.
43. Elabd, A.A., Elhefnawy, O.A. J. Fluorescence. 2016, 26, 271-276.
44. Abolghasemi, M.M., Sobhi, M., Piryaei, M. Sensors Actuat. B. 2016,

224, 391-395.
45. Rouis, A., Darbost, U., Bonnamour, I., Ouada, H.B. Mat. Chem. Phys.
2015, 164, 145-149.
46. Wang, Y., Wu, F. Polymer. 2015, 56, 223-228.
47. Coelho, L., Viegas, D., Santos, J.L., Marques Martins de Almedia, J.M.
Talanta. 2016, 148, 170-176.
48. Mishra, V., Patel, P.N., Tiwari, V. Opt. Quant. Electron. 2015, 47,
2299-2310.
49. Hogan, H. Photonics Spec. 2007, 41, 30.
50. Chamkouri, N., Niazi, A., Zare-Shahabadi, V. Spectrochim. Acta. Part

A. 2016, 156, 105-111.
51. Timbo, A.P., Pinto, P.V.F., Pinho, H.A., de Moura, L.P., Chretien, J.B.,
Viana, F.W., Filho, R.G.D., da Silva, E.B., da Silva, M.E.R., Menezes,
J.W.M., Guimaraes, G.F., Fraga, B.W. Sensors Actuat. B. 2016, 223, 406-
410.
52. Chu, C.S., Lin, K.Z., Tang, Y.H. Sensors Actuat. B. 2016, 223, 606-612.
53. Szili, M., Kasik, Matejec, V., Nagy, G., Kovacs, B. Sensors Actuat. B.
2014, 192, 92-98.
54. Zhao, L., Li, M., Liu, M., Zhang, Y., Wu, C., Zhang, Y. J. Hazard Mater.
2016, 15, 233-241.
38
55. Chen, Y., Wu, Y., Weng, B., Wang, B., Li, C. Sensors Actuat. B. 2016,
223, 689-696.
56. del Valle, M. Electroanal. 2010, 22, 1539-1555.
57. Kimmel, D.W., LeBlanc, G., Meschievitz, M.E., Cliffel, D.E. Anal. Chem.
2010, 84, 685-707.
58. Bratov, A., Abramova, N., Ipatov, A. Anal. Chim. Acta. 2010, 678, 149-
159.
59. Toko, K. Sci. Eng. C. 1996, 4, 69-82.
60. Toko, K. Sens. Update. 1998, 3, 131-160.
61. Major, N., Markovic, K., Krpan, M., Saric, G., Hruskar, M., Vahcic, N.
Talanta. 2011, 85, 569-574.
62. Escrichea, I., Kadar, M., Domenech, E., Gil-Snachez, L. J. Food Eng.
2012, 109, 449-456.
63. Kraujalyte, V., Venskutonis, P.R., Pukalskas, A., Cesoniene, L.,

Daubaras, R. Food Chem. 2015, 188, 583-590.
64. Ciosek, P., Augustyniak, E., Wroblebwski, W. Analyst. 2004, 129, 639-
644.
65. Paolesse, R., Lvova, L., Nardis, S., Di Natale, C., D Amico, A., Lo
Castro, F. Microchim. Acta. 2008, 163, 103-112.
66. Costa, A.M., Sobral, M.M.C., Delgadillo, I., Cerdeira, A., Rudnitskaya,
A. Sensors Actuat. B. 2015, 207, 1095-1103.
67. Hruskar, M., Major, N., Krpan, M., Vahcic, N. Talanta. 2010, 82, 1292-
1297.
68. Hruskar, M., Major, N., Krpan, M. Talanta, 2010, 81, 398-403.
39
69. Buratti, S., Sinelli, N., Bertone, E., Venturello, A., Casiraghi, E.,
Geobaldo, F. J. Food Agriculture. 2015, 95, 2192-2200.
70. He, W., Hu, X., Zhao, L., Liao, X., Zhang, Y., Zhang, M., Wu, J. Food
Research Int. 2009, 42, 1462-1467.
71. Volpati, D., Aoki, P.H.B., Dantas, C.A.R., Paulovich, F.V., de Oliveira,
M.C.F., Oliveira, O.N., Riul, A., Aroca, R.F., Constantino, C.J.L. Langmuir.
2012, 28, 1029-1040.
72. Bakas, I., hayat, A., Piletsky, S., Piletska, E., Chehimi, M.M., Noguer,
T., Rouillon, R. Talanta. 2014, 130, 294-298.
73. Teka, S., Gaied, A., Jaballah, N., Xiaonan, S., Majdoub, M. Mat.
Research Bull. 2016, 74, 248-257.
74. Pioggia, G., Di Francesco, F., Marchetti, A., Ferro, M., Leardi, R.,
Ahluwalia, A. Biosens. Bioelectron. 2007, 22, 2624-2628.
75. Pigani, L., Culetu, A., Ulrici, A., Foca, G., Vignali, M., Seeber, R. Food
Chem. 2011, 129, 226-233.
76. Czolkos, I., Dock, E., Tonning, E., Christensen, J., Winther-Nielsen, M.,
Carlsson, C., Mojzikova, R., Skladal, P., Wollenberger, U., Norgaard, L.,
Ruzgas, T., Emneus, J. Biosens. Bioelectron. 2016, 75, 375-382.
77. Granato, D., Margraf, T., Brotzakis, I., Capuano, E., van Ruth, S.M. J.
Food Sci. 2016, 80, C55-C65.
78. Campos, I., Bataller, R., Armero, R., Gandia, J.M., Soto, J., Martinez-
Manez, R., Gil-Sanchez, L. Food Res. Int. 2013, 54, 1369-1375.
79. Medina-Plaza, C., de Saja, J. A., Rodriguez-Mendez, M. L. Biosens.

Bioelectron. 2014, 57, 276-283.
80. Gutierrez-Capitan, M., Capdevilla, F., Vila-Plaza, J., Domingo, C.,

Buttgenbach, S., llobera, A., Puig-Pujol, A., Jimenez-Jorquera, C. J.
Sensors. 2014, 598317.
40
81. Ceto, X., Gonzalez-Calabuig, A., Capdevilla, J., Puig-Pujol, A., del
Valle, M. Sensors Actuat. B. 2015, 207, 1053-1059.
82. Blanco, C.A., de la Fuente, R., Caballero, I., Rodriguez-Mendez, M.L. J.

Food Eng. 2015, 157, 57-62.
83. Rodriguez-Mendez, M.L. Electronic Noses and Tongues in Food

Science. 2016. Ed. Elsevier. Philadelphia (USA).
84. Arrieta, A.A., Apetrei, C., Rodriguez-Mendez, M.L., de Saja, J.A.

85. de As, A.C., Cipri, A., Gonzalez-Calabuig, A., Stradiotto, N.R., del
Valle, M. Sensors Actuat. B. 2016, 222, 645-653.
86. Bagnsaco, L., Cosulich, M.E., Speranza, G., Medini, L., Oliveri, P.,
Lanteri, S. Food Chem. 2014, 157, 421-428.
87. Ghosh, A., Bag, A.K., Sharma, P., Tudu, B., Sabhapondit, S., Baruah,
B.D., Tamuly, P., Bhattacharyya, N., Bandyopadhyay, R. IEE Sensors J.
2015, 11, 6255-6262.
88. Wang, L., Niu, Q.F., Hui, Y.B., Jin, H.L. Sensors. 2015, 7, 17767-17785.
89. Rodriguez-Mendez, M.L., Parra, V., Apetrei, C., Villanueva, S., Gay,
M., Prieto, N., Martinez, J., de Saja, J.A. Microchim. Acta. 2008, 163, 23-
31.
90. Garcia-Hernández, C., Garcia-Cabezón, C., Medina-Plaza, C., Martin-

Pedrosa, F., Blanco, Y., de Saja, J.A., Rodriguez-Mendez, M.L. Beilstein J.
Nanotechnol. 2015, 6, 2052-2061.
91. Medina-Plaza, C., Garcia-Cabezon, C., Garcia-Hernandez, C.,

Bramorski, C., Blanco-Val, Y., Martin-Pedrosa, F., Kawai, T., de Saja, J.A.,
Rodriguez-Mendez, M.L. Anal. Chim. Acta. 2015, 853, 572-578.
92. Alex, S., Tiwari, A. J. Nanosc. Nanotech. 2015, 15, 1869-1894.
93. Lim, W.Q., Gao, Z.Q. Electroanal. 2015, 27, 2074-2090.
41
94. Jiang, J. Functional Phthalocyanine Molecular Materials. Series:

Structure and Bondings. 2010. Ed. Springer. Berlin (Germany)
95. Zagal, J.H., Griveau, S., Silva, J.F., Nyokong, T., Bedioui, F. Coord.
Chem. Ver. 2010, 254, 2755-2791.
96. Rodriguez-Mendez, M.L. Sensing properties of phthalocyanines in

Encyclopedia of sensors. 2006, 111-134. Eds. Grimes, C.S., Dickey, E.C.,
Pishko, M.V. Ed. America Scientific Publishers. California (USA).
97. Palmer, T., Bonner, P.L. Enzymes: Biochemistry, Biotechnology,

Clinical Chemistry. 2007. Ed. Woodhead Publishing. New Delhi (India).
98. Yamada, Y., Aida, K., Uemura, T. Agr. Biol. Chem. 1966, 30, 95-96.
99. Zhou, J., Li, X., Yang, L.L., Yan, S.L., Wang, M.M., Cheng, D., Chen, Q.,
Dong, Y.L., Liu, P., Cai, W.Q., Zhang, C.C. Anal. Chim. Acta. 2015, 899, 57-
65.
100. Wang, W.J., Gao, J., Wang, L., Ye, B.X. Anal. Methods. 2015, 7,
8847-8856.
101. Medina-Plaza, C., García-Hernández, C., de Saja, J.A., Fernández-

Escudero, J.A., Barajas, E., Medrano,G., García-Cabezón,C., Martin-
Pedrosa, F., Rodriguez-Mendez, M.L. LWT-Food Sci. Tech. 2015, 62, 940-
947.
102. Mukdasai, S., Crowley, U., Pravda, M., He, X.Y., Nesterenko, E.P.,
Nesterenko, P.N., Paull, B., Srijaranai, S., Glennon, J.D., Moore, E.
Sensors Actuat. B. 2015, 218, 280-288.
103. Baghizadeh, A., Karimi-Maleh, H., Khoshnama, Z., Hassankhani, A.,

Abbasghorbani, M. Food Anal. Methods. 2015, 8, 549-557.
104. Gay-Martin, M., Diez-Arevalo, E., Rodriguez-Mendez, M.L., de Saja,

J.A. Sensors Actuat. B. 2013, 185, 24-31.
105. www.dropsens.com
42
106. Li, Y.F., Ye, Z., Luo, P.L., Li, Y., Ye, B.X. Anal. Methods. 2014, 6, 1928-
1935.
107. Raoof, J.B., Ojani, R., Baghayeri, M. Turkish J. Chem. 2013, 37, 36-
50.
108. Medina-Plaza, C., Furini, L.N., Constantino, C.J.L., de Saja, J.A.,

109. Zou, L.N., Li, Y.F., Cao, S.K., Ye, B.X. Talanta. 2013, 117, 333-337.
43
CHAPTER 2
Analysis of Organic Acids and Phenols of Interest
in the Wine Industry using Langmuir–Blodgett
Films Based on Functionalized Nanoparticles
Chapter 2: Introduction
As stated on the introduction, antioxidants (i.e. phenols) present great

interest for the enology industry due not only to their influence on the
organoleptic characteristics of the final product but also due to their
health benefits. Free-radical theory of aging states that organisms age
because cells tend to accumulate free radical damage over time. In some
model organism, such as yeast, there are evidences that reducing
oxidative damage can extend life span. It has been demonstrated that
antioxidants are helpful in reducing and preventing damage from free
radical reactions because of their ability to neutralize them. Although a
variety of methods have been described to detect antioxidants and to
evaluate the antioxidant capacity of foods, up to now, a single method
has not been recognized as the most adequate. Antioxidants are usually
electroactive compounds and this property can be exploited for
electrochemical detection.
In addition, organic acids present in grapes and wines need to be

monitored as they are related to the pH, acidity and mouthfeel among
other properties of the wine. They also influence the stability of the final
product and “control” the microbiological modifications that can occur
along the winemaking process.
In this chapter ITO modified with gold nanoparticles deposited by

Langmuir-Blodgett technique are used to analyze standard solution of
phenols (caffeic and gallic acid) and organic acids (tartaric, malic, citric
and lactic acids) present in musts. The modification of the electrode
using LB technique allow preparing thin films well organized and
structured, increasing the number of active sites on the surface and
improving the analytical properties of the sensor. The electrocatalytic
properties, the dynamic character and the detection limits of these
amphiphilic gold nanoparticles modified with n-dodecanethiol will be
evaluated. Moreover, mixtures of tartaric and caffeic acids will be
analyze in order to detect possible interferences during the analysis. This
research has been developed in collaboration with Prof. Kawai from
Tokyo University of Science (Japan) and has been published in Analytica
Chimica Acta.
47
2.1 INTRODUCTION
The organoleptic and antioxidant properties of grapes and wines are
closely related to their chemical composition. The organic acid content
of grapes, musts and wines is of interest because its important influence
on their sensory properties such as flavor, taste, color and aroma.
Organic acids also affect the juice stability and are used as indicators of
microbiological alterations in the beverage1,2. The principal organic acids
found in grapes and therefore in wines, are tartaric and malic acids.
Other acids such as citric acid are present in smaller amounts3. During
the course of winemaking and in the finished wines other acids such as
lactic acid can play significant roles.
Grapes and wines are also rich in phenolic compounds. Such compounds
have attracted great interest due to their antioxidant activity and their
important influence in the organoleptic properties4,5. In particular,
phenolic acids, such as gallic and caffeic acids have been studied for
their antioxidant capacity and for acting as venous dilators6.
Various analytical methods have been reported for the determination of

organic acids and phenolic compounds in wines and grapes including gas
chromatography, HPLC, spectroscopic or electrochemical techniques7-9.
Electrochemical measurements have certain advantages for the
determination of antioxidants. For instance, the oxidation potentials
measured by cyclic voltammetry (CV) have been used to compare the
antioxidant strength of compounds such as phenolic acids, flavonoids,
cinnamic acids, etc. being the glassy carbon electrode (GCE) the more
frequently used electrode10-14. Chemically modified electrodes have also
been successfully used as sensitive and selective tools for determination
of many organic substances and some of them have been dedicated to
the detection of organic acids and/or phenolic compounds15-18.
Voltammetric electrodes have been successfully integrated in
multisensory systems for the analysis of complex liquids19,20.
Gold nanoparticles (AuNPs), have received great interest in the field of

electrochemical sensors21. Their small size provides high surface to
volume ratios which are the reason of their unique electronic and
48
catalytic properties22-26. However, in a number of applications, the

benefits of using nanoparticles have not been fully exploited due to the
lack of strategies for positioning the particles in ordered, homogeneous
and reproducible solid-state devices. Recently, attempts have been
made to immobilize NPs onto solid substrates using the Langmuir-
Blodgett (LB) technique27-30. In order to obtain such ordered monolayers,
nanoparticles must be capped to induce the amphiphilic character
necessary for the LB technique31.
2.2 MATERIALS AND METHODS
2.2.1- Chemicals
All experiments were carried out in deionized Milli-Q water (Millipore,
Bedford, MA). Inorganic salts, organic acids (DL-malic acid, L(+)-tartaric
acid and lactic acid), phenolic acids (caffeic and gallic acids),
dodecanethiol, tetrachloroaurate tetrahydrate, tetraoctylammonium
bromide and NaBH4 were purchased from Sigma-Aldrich. Commercially
available reagents and solvents were used without further purification.
10-3 mol·L-1 stock solutions of organic acids and phenolic compounds

were prepared by solving the corresponding compound in KCl (0.1 mol·
L-1). Solutions with lower concentration were prepared from the stock
solutions by dilution. The mixtures phenol/organic acid were prepared
by mixing the corresponding volume of the 10-3 mol·L-1 stock solutions in
proportions 40:10, 25:25 and 10:40 to obtaining 50 mL as final volume.
The final concentrations of the mixtures were 810-4:210-4,

510-4:510-4 and 210-4:810-4 mol·L-1 respectively.
2.2.2- Equipment
LB films were prepared in a USI-3-22 film balance (USI Co. Ltd.) equipped
with a Wilhelmy plate to measure the surface pressure. TEM Images
were obtained using a Jeol JEM-1011 electron microscope operated at
an accelerating voltage of 100 kV. The electrochemistry was carried out
in an EG&G PARC 273 potentiostat/galvanostat.
49
Chapter 2: Materials and Methods
2.2.3- Preparation of nanoparticles

Dodecanethiol-functionalized Au nanoparticles (SDODAuNP) were
prepared using the Brust's two-phase method32. A quantity 15.8 mL of a
24.310-3 mol·L-1 aqueous solution of HAuCl4 was added to a 2.2910-3
mol·L-1 phase transfer reagent, tetraoctylammonium bromide, dissolved
in toluene (15 mL), and the mixture was stirred for 10 min. The toluene
phase was subsequently collected and 1.5 mL of a 4910-3 mol·L-1
toluene solution of dodenanethiol (DOD) was added. The solution was
stirred for 2 h. Then, 2 mL of a 2.11 mol·L-1 aqueous solution of NaBH4
was added. The reaction mixture was vigorously stirred at room
temperature for 12 h and the toluene phase was collected. Since the size
distribution of the resulting SDODAuNP was quite broad, the dispersion
was refluxed at 80oC for 24 h to obtain monodisperse SDODAuNP. The
resulting SDODAuNP were washed five times with acetone to remove the
phase transfer reagent, excess DOD and reaction byproducts. The
resulting particles have a narrow size distribution with an average
diameter of 4.90.5 nm.
2.2.4- Preparation of LB films

The toluene dispersion of SDODAuNP was spread onto the air/water
interface to make the monolayer of AuNPs. The surface pressure-area
(π-A) isotherm was measured at room temperture using a Wilhelmy
plate. The monolayer of SDODAuNP was then transferred onto ITO
substrates at a surface pressure of 10 mN·m-1 to obtain LB films.
2.2.5- Electrochemical studies

The LB films were used as working electrodes of a conventional three
electrode cell. The reference electrode was Ag|AgCl/KCl 3M and the
counter electrode was a platinum plate. Cyclic voltammetry was carried
out at room temperature with a scan rate of 0.1 V·s−1 in the potential
range of -1.0 V and +1.0 V.
The repeatability of the voltammograms was evaluated from 5

repetitions on each sample. The reproducibility of data provided by the
SDODAuNP was evaluated by comparing data provided by three sensors
measuring identical samples in different days.
50
The dynamic behavior of the films was characterized by registering

voltammograms at increasing scan rates from 0.010 to 1.0 V s-1.
The calibration curves were constructed from solutions with

concentrations ranging from 110-3 to 110-5 mol·L-1.
The limits of detection (LOD) were calculated following the 3sd/m

criterion, where m is the slope of the calibration graph, and sd was
estimated as the standard deviation (n=5) of the voltammetric signals at
the concentration level corresponding to the lowest concentration of
the calibration plot33.
2.3 RESULTS AND DISCUSSION
2.3.1- Structural characterization

Figure 2.1 shows a TEM image of LB monolayer of SDODAuNP transferred
onto a TEM-copper grid at a surface pressure of 10 mN·m-1, which is the
same transfer condition used to prepare SDODAuNP monolayer onto ITO
substrates. SDODAuNP were in a relatively random array, but the single
layer was densely covered with the substrate. The diameter of SDODAuNP
was 4.90.5 nm. The density of SDODAuNP was evaluated from the TEM
image and found to be ~30 nm2/particle.
Figure 2.1. TEM image of SDODAuNP LB monolayer transferred onto a TEM-

copper grid.
51
Chapter 2: Results and Discussion
2.3.2- Electrochemical response in simple electrolytes

As a first electrochemical characterization, cyclic voltammograms of
bare ITO and of SDODAuNP-LB on ITO immersed in 0.1 mol·L-1 solutions of
different electrolytes (KCl, KBr, KNO3, KClO4 and MgCl2) were registered.
Figure 2.2 compares the voltammograms obtained from a bare ITO and
from SDODAuNP-LB electrode in KCl and MgCl2 solutions over a potential
range from -1.0 to +1.0 V.
Figure 2.2. Cyclic voltammograms of bare ITO (blue line) and SDODAuNP-LB (red
-1 -1
line) sensors immersed in a) 0.1 mol·L KCl and b) 0.1 mol·L MgCl2. Scan rate,
−1
0.1 V·s .
For the bare ITO electrode, no observable faradaic current appeared

either on the forward or reverse scans regardless the electrolyte used.
Whereas, the cyclic voltammogram for the SDODAuNP-LB modified
electrode immersed in KCl (Figure 2.2.a) showed the typical response of
gold nanoparticles with the oxidative peak at 0.7 V corresponding to
gold oxide formation during the forward scan alongside the reduction
peak at 0.17 V responsible for the reduction of gold oxide during the
reverse scan32,33. The reduction of protons from water was promoted in
the presence of SDODAuNP as indicated by the presence of a cathodic
peak at ca. -0.75 V that was absent on neat ITO glass.
Voltammograms registered in the presence of other electrolytes

containing potassium ion (KBr, KNO3 or KClO4) were quite similar. The
peak positions did not change and only small modifications in the redox
52
peak currents were observed. However, in the presence of a divalent

cation (such as Mg2+ from MgCl2) the intensity of the redox peaks
corresponding to gold oxidation/reduction decreased. Simultaneously,
the peak corresponding to the reduction of protons from water was
displaced to more negative potentials (Figure 2.2.b).
From these results, it can be concluded that cations instead of anions

diffuse inside the films to maintain the electroneutrality during the scan.
In addition, when the cation radius increased, the redox peak potential
of the reduction of protons from water moved to more negative
potentials decreasing the electrocatalytic activity of the SDODAuNP-LB
sensor.
In all electrolytes analyzed, the intensity of the peaks increased linearly

with the scan rate pointing to an electron transfer limited process (due
to the electrochemical activity of the nanoparticles deposited at the
surface of the electrode) with regression coefficients in the range 0.990-
0.995.
2.3.3- Electrocatalytic effect of SDODAuNP-LB towards organic

acids
SDODAuNP-LB modified electrodes were used to analyze organic acids
typically found in grapes and wines. The study included the two main
organic acids present in wines (tartaric and malic acids), one acid
present in minor proportion (citric acid) -all of them coming from the
oxidation of sugars-, along with lactic acid which is an example of acid
generated by microbial activity. The electrocatalytic effect of SDODAuNP-
LB electrodes was analyzed in 10-3 mol·L-1 solutions of the corresponding
organic acid by comparing the cyclic voltammograms observed at the
bare ITO electrode and at the SDODAuNP-LB electrodes (Figure 2.3).
None of the organic acids studied produced noticeable electrochemical

responses when analyzed with a bare ITO electrode. However, the
modified electrode catalyzed the reduction of the liberated protons and
intense oxidation/reduction processes were observed at negative
potentials.
53
The different electrochemical behaviors were related to the chemical

nature of the acids analyzed. The voltammetric response of lactic acid
(monoprotic acid, pKa=3.90, pH of a 10-3 mol·L-1 solution=3.43) showed
only one reduction peak at -0.85 V that corresponds to the reduction of
the dissociated H+ proton. In malic acid (diprotic acid, pK1=3.48 and
pK2=5.10, pH of a 10-3 mol·L-1 solution=3.18), two well-separated
electrochemical reduction peaks for each one of the dissociated protons
were observed at about −0.85 and −0.95 V. The corresponding
oxidations occurred at -0.4 V and -0.5 V. The response of tartaric acid
(diprotic acid, pKa1=2.98 and pKa2=4.34, pH of a 10-3 mol·L-1
solution=2.93) also showed the reduction of the two protons (-0.85 V
and -0.9 V). The oxidations were observed at -0.4 V and -0.5 V. It is
important to notice that in tartaric acid, the redox peaks were not well
resolved in both the forward and the reverse scans. It is also interesting
to remark that the Ia/Ic ratio was much higher for tartaric acid than for
malic acid. The fact that tartaric and malic acids showed different
responses is of particular interest for the wine industry since tartaric and
malic acids are the most prevalent acids present in wines. Moreover,
malic acid shows a continuous decrease during ripening whereas tartaric
acid remains almost unchanged. Therefore, different ratios can be
obtained during ripening and the optimum harvest date can be
established from their ratio. The response of citric acid (triprotic acid,
pK1=3.09, pK2=4.75 and pKa3=6.41, pH of a 10-3 mol·L-1 solution=2.97)
was similar to that of tartaric acid and two not well resolved cathodic
peaks were observed. The similitude could be easily explained taking
into account the weak acidity of the third acidic proton of the citric acid.
According to the enhancement in current responses, it is possible to

conclude that SDODAuNP-LB electrodes show an intense electrocatalytic
effect towards the reduction of protons, and that voltammetric
responses are different for the main acids found in wines. In addition,
the observation of distinct redox processes associated with first and
second dissociated protons is a remarkable result of our SDODAuNP-LB
electrodes since bulk gold electrodes are not able to discriminate among
acids or to provide distinct signals for different protons36.
54
-3
Figure 2.3. CV of a bare ITO (blue line) and SDODAuNP-LB (red line) towards 10
-1 -1
mol·L of a) Lactic, b) Malic, c) Tartaric and d) Citric acids. Scan rate, 0.1 V·s .
The influence of the potential scan rate on the peak height was also
investigated between 0.01 and 1.0 V·s-1. The intensity of the cathodic
peaks increased linearly with the square root of the scan rate and
conformed to the Randles-Sevcik equation (Equation 2.1):
I=2.687×105 n3/21/2 D1/2 AC (Equation 2.1)
Where I is the peak current (A), C represents concentration of the

electroactive species (mol·L-1),  potential scan rate (V·s-1), A electrode
surface (cm2), D diffusion coefficient of the analytes (cm2·s-1), and n
number of electrons transferred in the redox process. In all cases, the
correlation coefficient R2 was higher than 0.990, pointing to a diffusion
limited process (Table 2.1).
The logarithm of peak current changed linearly with the logarithm of

scan rate with a slope value close to 0.5 for tartaric and lactic acids
55
confirming ideal diffusion controlled mechanism. In the case of citric and

malic acids, the slopes were found to be lower than 0.5. A closer look to
the graphs showed that the increase in the scan rate caused a shift of
the cathodic peaks towards more negative potentials. At scan rates
higher than 0.05 V·s-1, the peaks could not be longer observed. The
slopes recalculated for citric and malic acids using scan rates from 0.01
to 0.05 V·s-1 were close to 0.4 pointing also to a diffusion controlled
process.
Table 2.1. Slope and regression coefficients obtained from the voltammograms
registered at different scan rates for the organic acids. Intensities measured
using the cathodic wave at -0.85V.
Relationship with scan rate Relationship with scan rate

Organic log Ic (A) vs. log 𝑣 (mV·s-1)
Ic (A) vs. 𝑣1/2 (mV·s-1)1/2
acid
Slope R2 Slope R2
Lactic -24.14 0.990 0.48 0.990
Malic -9.39 0.980 0.24 0.982
Tartaric -47.15 0.992 0.45 0.991
Citric -22.69 0.981 0.26 0.982
In contrast, the anodic wave showed a tricky behavior. At low scan rates,
a linear dependence with the square root of the scan rate was observed,
but at scan rates higher than 0.5 V·s-1 the voltammograms presented
distorted shapes and decreased their intensity while a new broad peak
at 0.5 V grew progressively. These facts could be explained assuming
that adsorption/polymerization processes are involved in the oxidations,
causing the mentioned distortion37.
The effect of the concentration in the sensor response was studied by

immersing the electrode in organic acid solutions with concentrations
ranging from 110-5 to 110-3 mol·L-1. The influence of increasing
concentrations is illustrated in Figure 2.4 for citric acid. When
representing the calibration curve for the cathodic peak (at -0.85 V), an
increase in the intensity of the peaks with the concentration was
56
observed. The limit of detection (LOD) calculated following the 3sd/m

criterion was 5.610-6 mol·L-1 (Table 2.2). The anodic peak was only
observed at concentrations higher than 10-4 mol·L-1. This observation
could be explained taking into account the dissociation constants of the
acids (pKa) and the change of the pH caused by the concentration. The
pH of a 10-3 mol·L-1 citric acid solution (CitH3) is 2.97 and the
predominant species is CitH2- whereas the pH of the 10-5 mol·L-1 solution
is 5.01 and the predominant species is CitH2- so, the anodic wave could
only be observed when the predominant specie was CitH2-. A similar
behavior was observed in the rest of organic acids studied. For instance,
tartaric acid can be present in wine and juice as tartaric acid (TH2), bi-
tartrate (TH-) or tartrate (T2-). The ratio of these species depends mainly
on the pH of the wine. The anodic wave was only observed when the
predominant species was bitartrate TH- (maximum concentration of TH-
occurs at pH 3.7 and at pH 5.0, the predominant species is T2-).
Due to the non-linear behavior of the anodic wave, in all organic acids
analyzed the LODs were calculated by analyzing the intensity of the
cathodic peaks at -0.85 V following the 3sd/m criterion. As observed in
Table 2.2, they were in the range of 10-5-10-6 mol·L-1.
Figure 2.4. CV of a LB film of SDODAuNP-LB immersed in increasing amounts of

-5 -3 -1
citric acid (10 to 10 mol·L ).
57
2
Table 2.2. LOD, sensitivity and regression coefficient (R ) calculated from the
cathodic peak at ca. -0.85V for the four organic acids under study.
Organic acid LOD (mol·L-1) Slope R2

-6
Lactic 17.63 10 -0.19 0.980
Malic 5.91 10-6 -0.53 0.999
Tartaric 4.73 10-6 -0.66 0.990
Citric 5.62 10-6 -0.56 0.999
Taking into account that in the must obtained from grapes, tartaric acid
is found in the range 3–7 g·L−1 and malic acid in the range 1–3 g·L−1 38 the
linear range and the LOD found here is adequate for the concentration
usually found in such products.
2.3.4- Electrocatalytic effect of SDODAuNP-LB towards phenolic

acids
Phenolic acids are also important ingredients of grapes and wines. Their
interest relies on their antioxidant activity and on the key role they play
in the organoleptic characteristics of wines. The electrochemical
techniques are also appropriate to evaluate the antioxidant (or
electrochemical) activity of the phenols. It has been established that the
redox activity of these compounds can be measured using a glassy
carbon electrode10,15,39. Depending on their chemical structure phenolic
compounds can show low oxidation potentials (at ca. 0.5 V) or high
oxidation potential (at ca. 0.8 V). In this chapter, the electrocatalytic
effect of SDODAuNP-LB on the electrochemical response of two phenolic
acids commonly found in wines, one with low oxidation potential
(caffeic acid) and one with high oxidation potential (gallic acid) is
analyzed (Figure 2.5).
The electrochemical response of caffeic acid at a bare ITO was

characterized by a redox pair with the anodic wave at 0.55 V and the
cathodic at 0.25 V assigned to the oxidation/reduction of the diphenol to
the o-quinone (Figure 2.5.a). The peak current of the electro-oxidation
of caffeic acid at the SDODAuNP-LB sensor was clearly enhanced from 170
A to 270 A (an increase of the 75%) evidencing the catalytic effect of
58
the modified electrode. The electrocatalytic effect was even more

pronounced in gallic acid (a triphenol of high oxidation potential). As
shown in Figure 2.5.b the gallic acid could not be oxidized nor reduced
on a bare ITO glass at the potential range from -1.0 V to +1.0 V.
However, at the modified electrode, an intense and irreversible peak
appeared at 0.75 V. The enhancement in current response was a clear
evidence of the catalytic effect of the SDODAuNP-LB modified electrode
towards the oxidation of gallic acid.
Figure 2.5. Voltammetric response at a bare ITO electrode (blue line) and
-4
SDODAuNPs-LB electrode (red line) towards a) caffeic acid and b) gallic acid 10
-1
mol·L .
The influence of the potential sweep speed on the peak height was
studied in 10-3 mol·L−1 solutions. The scan rate varied between 0.01 and
1.0 V·s-1. A linear increment in the redox peak currents as a function of
the square root of the scan rate was observed (with R2 of 0.990)
indicative of a diffusion-controlled electrode process. The correlation
between the logarithm of peak current and the logarithm of scan rate
was linear and slopes were close to 0.5 (0.55 for caffeic acid, 0.42 for
gallic acid). Thus, the diffusion controlled mechanism was further
confirmed.
Voltammograms were registered using phenolic solutions with different

concentrations. A linear relation between the peak currents and caffeic
acid concentration in the range between 110-3 mol·L−1 and 1 10-5
59
mol·L−1 was obtained. The LOD calculated was 0.6710-6·mol·L-1 and

1.2110-6 mol·L-1 (for the anodic and cathodic peak respectively).
Simultaneously, the pH variation produced a progressive shift of the
redox potential to higher values. This in good agreement with previously
published studies that have proved that pH affects the redox potentials
of the phenols10. CV plots recorded in gallic acid showed a similar
behavior. A clear oxidation peak at ca. 0.75 V was observed that
increased its intensity with the concentration (Figure 2.6). The obtained
calibration graph showed a linear dependence between peak height and
gallic acid concentration (oxidation: R2=0.997 / reduction: R2=0.997). The
LOD calculated according with the 3Sd/m criterion was 0.5110-6 mol·L-1.
This LOD obtained using the SDODAuNP-LB sensors was much lower than
the values reported for glassy carbon electrodes (ca. 10-4 mol·L-1)10 and
similar to the LOD obtained using electrocatalytic LB films prepared from
organic compounds38. It is also important to notice that the linear range
was clearly enhanced with respect to that observed in glassy carbon
electrodes, indicating that the contamination of the electrodes by the
oxidation products is less important41.
Figure 2.6. a) Cyclic voltammograms obtained for different caffeic acid

-5 -3 -1
concentrations from 10 to 10 mol·L . b) Calibration graph for the
determination of caffeic acid, ( reduction peak,  oxidation peak).
60
2.3.5- Response of SDODAuNP-LB towards mixtures of organic

acids and phenols
In this section, experiments were made varying the acidity of the
solution using mixtures of caffeic acid and tartaric acid (the majoritarian
acid in wine and the main responsible of its acidity) in the range of pHs
and concentrations found in wines. Experiments were carried out in
caffeic:tartaric mixtures 40:10, 25:25 and 10:40 (corresponding to 8
10-4:210-4, 510-4:510-4, 210-4:810-4 mol·L-1 respectively). Figure 2.7
shows the voltammetric responses of mixtures of caffeic and tartaric
acid analyzed using the SDODAuNP-LB electrodes.
As expected, the intensity of both oxidation and reduction peaks, Ia and

Ic, of caffeic acid increased linearly with the concentration. At low
concentrations of tartaric acid, the only peaks observed were produced
by the oxidation/reduction of caffeic acid. Increasing amounts of tartaric
acid decreased progressively the pH and the redox activity of the
dissociated protons could be observed at negative potentials. Tartaric
acid and caffeic acid did not show important interferences in the studied
range, except a slight shift in the positions of caffeic acid peak due to the
change in the pH caused by tartaric acid. The LOD obtained were similar
to those obtained separately confirming the possibility of using the
SDODAuNP-LB sensor for the simultaneous determination of acidity and
caffeic acid in the range of concentrations usually found in wines. Similar

results were obtained for gallic acid.
61
Figure 2.7 CV registered using SDODAuNP-LB electrode immersed in

caffeic:tartaric mixtures: 40:10 (green line), 25:25 (blue line), 10:40 (red line).
2.3.6- Electrode repeatability and reproducibility

In order to characterize the repeatability of the measurements
registered with the SDODAuNP-LB electrode, repetitive measurements
were carried out in 10-3 mol·L-1 solutions. For all the samples analyzed,
the results of 5 consecutive measurements showed a relative standard
deviation (RSD) lower than 2 %.
Also the reproducibility of the electrodes was examined by the

determination of 10-3 mol·L-1 caffeic acid using three electrodes
prepared using the same method. RSD of anodic peak potential was
found to be less than 4%, confirming the reproducibility of the method.
2.4 CONCLUSIONS
In this chapter, electrodes chemically modified with functionalized gold
nanoparticles were achieved by Langmuir-Blodgett technique. The
modified SDODAuNP-LB electrodes demonstrated efficient interfacial
properties being able to detect organic acids responsible of the acidity of
grapes, musts and wines and phenolic acids with antioxidant properties.
The LB technique applied to these amphiphilic nanoparticles produced
sensors with high surface to volume ratios and efficient electrocatalytic
behavior (increased peak currents) with limits of detection much lower

than those obtained on bare ITO glass (in the range of 10-6 mol·L-1). In
addition, the SDODAuNP-LB sensors were able to discriminate the
predominant species indicated by the pKa and the pH. Finally, the
simultaneous detection of organic acids and phenols was possible due to
the lack of interferences in the pH and concentrations rage found in
wines and musts.
62
Chapter 2: Conclusions
References
1. Ali, K., Maltese, F., Choi, Y.H., Verpoorte, R. Phytochem. Rev. 2010, 9,
357-378.
2. Lima, M.S., Silani, I.S.V., Toaldo, I.M., Correa, L.C., Biasoto, A.C.T.,
Pereira, G.E., Bordignon-Luiz, M.T., Ninow, J.L. Food Chem. 2014, 161,
94-103.
3. Soyer, Y., Koca, N., Karadeniz, F. J. Food Comp. Anal. 2003, 16, 629-
636.
4. Vauzour, D., Rodriguez-Mateos, A., Corona, G., Oruna-Concha, M.J.,

Spencer, J.P.E. Nutrients. 2010, 2, 1106-1131.
63
5. Teissedre, P.L., Jourdes, M. Tannins and Anthocyanins of Wine:

Phytochemistry and Organoleptic Properties, in Natural Products, 2013,
2255-2274. Ed. Springer, Berlin (Germany).
6. Mudnic, I., Modun, D., Rastija, V., Vukovic, J., Brizic, I., Katalinic, V.,
Kozina, B., Medic-Saric, M., Boban, M. Food Chem. 2010, 119, 1205-
1210.
7. Organisation Internationale de la Vigne et du Vin, "Recueil des

methods internationals d'analyse des vins et des mouts, edition 2011",
8th Assemblée Générale, 21 June 2010, Paris, 2011.
8. Lee, J. Food Sci. Nutr. 2014, 2, 1-8.
9. Prehn, R., Gonzalo-Ruiz, J., Cortina-Puig, M. Current Anal. Chem. 2012,

8, 472-484.
10. Kilmartin, P.A., Zou, H.L., Waterhouse, A.L. J. Agric. Food Chem. 2001,
49, 1957-1965.
11. Bortolomeazzi, R., Sebastianutto, N., Toniolo, R., Pizzariello, A. Food

Chem. 2007, 100, 1481-1489.
12. Abou Samra, M., Chedea, V.S., Economou, A., Calokerinos, A.,
Kefalas, P. Food Chem. 2011, 125, 622-629.
13. Yakovleva, K.E., Kurzeev, S.A., Stepanova, E.V., Fedorova, T.V.,

Kuznetsov, B.A., Koroleva, O.V. Appl. Biochem. Microbiol. 2007, 43, 661-
668.
14. Arteaga, J.F., Ruiz-Montoya, M., Palma, A., Alonso-Garrido, G.,

Pintado, S., Rodriguez-Mellado, J.M. Molecules. 2012, 17, 5126-5138
15. Kilmartin, P.A., Zou, H.L., Waterhouse, A.L. Am. J. Enol. Vitic. 2002,
53, 294-302.
16. Schmitt, R.E., Molitor, H.R., Wu, T.S. Int. J. Electrochem. Sci. 2012, 7,
10835-10841.
64
17. Bianchini, C., Curulli, A., Pasquali,M., Zane, D. Food Chem. 2014, 156,
81-86.
18. Matemadombo, F., Apetrei, C., Nyokong, T., Rodríguez-Méndez,

M.L., de Saja, J.A. Sens. Actuators B. 2012, 166-167, 457-466.
19. Gay-Martin, M., de Saja, J.A., Muñoz, R., Rodriguez-Mendez, M.L.

20. Apetrei, I.M., M.L. Rodríguez-Méndez, M.L., Apetrei, C., Nevares, I.,
del Alamo, M., de Saja, J.A. Food Res. Int. 2012, 45, 244-249.
21. Daniel, M.C., Astruc, D. Chem. Rev. 2004, 104, 293-346.
22. Campbell, F.W., Compton, R.G. Anal. Bioanal. Chem. 2010, 396, 241-
259.
23 Pingarron, J.M., Yañez-Señedo, P., Gonzalez-Cortes, A. Electrochim.

Acta. 2008, 53, 5848-5866.
24. White, R.J., Luque, R., Budarin, V.L., Clark, J.H., Macquarrie, D.J.
Chem. Soc. Rev. 2009, 38, 481-494.
25. Fan, M.K., Andrade, G.F.S., Brolo, A.G. Anal. Chim. Acta. 2011, 693, 7-
25.
26. Baghbamidi, S.E., Beitollahi, H., Mohammadi, S.Z., Tajik, S., Soltani-
Nejad, S., Soltani-Nejad, V. Chin. J. Catal. 2013, 34, 1869-1875.
27. Wang, L., Chen, X.H., Wang, X.L., Han, X.P., Liu, S.F., Zhao, C.Z.
Biosens. Bioelectron. 2013, 30, 151-157.
28. Huang, X.J., Choi, Y.K. Sensors Actuat. B. 2007, 122, 659-671.
29. Sanders, T.A., Sauced, M.N., Dahl, J.A. Mater. Lett. 2014, 120, 159-
162.
30. Supian, F.L., Richardson, T.H., Nabok, A.V., Deasy, M., Syahriman, M.,
Azmi, M. Adv. Mat. 2014, 895, 520-525.
65
31. Capone, S., Manera, M.G., Taurino, A., Siciliano, P., Rella, R., Luby, S.,
Benkovicova, M., Siffalovic, P., Majkova, E. Langmuir. 2014, 30, 1190-
1197.
32. Brust, M., Fink, J., Bethell, D., Schiffrin, D.J., Kiely, C. J. Chem. Soc.,
Chem. Commun. 1995, 1655-1656.
33. Nomenclature. “Symbols, Units and Their Usage in Spectrochemical

Analysis−II Data Interpretation”. Analytical Chemistry Division, IUPAC
1978.
34. Alonso-Lomillo, M.A., Yardimci, C., Dominguez-Renedo, O., Arcos-

Martinez,M.J. Anal. Chim. Acta. 2009, 633, 51-56.
35. Fu, Y.Y., Liang, F.F., Tian, H.F., Hu, J.B. Electrochim. Acta. 2014, 120,
314-318.
36. Escobar, J.D., Alcaniz, M., Masot, R., Fuentes, A., Bataller, R., Soto, J.,
Barat, J.M. Food Chem. 2013, 138, 814-820.
37. Arteaga, J.F., Ruiz-Montoya, M., Palma, A., Alonso-Garrido, G.,

Pintado, S., Rodriguez-Mellado, J.M. Molecules. 2012, 17, 5126-5138.
38. Cabanis, J.C., Flanzy, C. Oenologie, Technique & Documentation,

1998, Ed. Lavoisier, Paris (France).
39. Apetrei, C., Apetrei, I., de Saja, J.A., Rodriguez-Méndez, M.L. Sensors.
2011, 11, 1328-1344.
40. Giuliani, E., Fernandes, R., Brazaca, L.C., Rodríguez-Méndez, M.L., de

Saja, J.A., Zucolotto, V. Biosens. Bioelectron. 2011, 26, 4715-4719.
41. Ferreira, M., Varela, H., Torresi, R.M., Tremiliosi-Filho, G.

66
67
CHAPTER 3
Structural and Electrochemical Properties of
Lutetium Bis-Octachloro-Phthalocyaninate
Nanostructured Films. Application as
Voltammetric Sensors
The modification of ITO using Langmuir-Blodgett (LB) technique has

shown to be very useful to get nanostructured sensors with high number
of active sites leading to an improvement on the electrochemical
analysis of antioxidants. Other technique to obtain this kind of
nanostructures is Langmuir-Schaeffer (LS) in which the transfer of the
thin films onto the substrate is made horizontally instead of vertically. In
this chapter, differences between sensors prepared by both techniques
will be discussed.
In this case, the electrocatalytic modifier is a phthalocyanine derivate

bis[2,3,9,10,16,17,23,24-octachlorophthalocyaninate] lutetium(III)
complex (LuPc2Cl32). As detailed on the introduction, phthalocyanines
are among the most suitable materials for electrochemical sensors due
to their versatility and to their unique electrochemical and
electrocatalytic properties. Electrochemical sensors based on
phthalocyanine derivates have been widely used to analyze food and
beverages, focusing on their antioxidant capacity.
Electrochemical and structural properties of voltammetric sensors

prepared using LB and LS will be analyzed in this chapter. Their
application as sensing units for detecting catechol (a polyphenol model
molecule) will be discussed on the basis of their stability and limits of
detection.
This work has been published in Journal of Nanoscience and

Nanotechnology and was carried out in collaboration with researchers
from UNESP (Brazil).
3.1 INTRODUCTION
Phthalocyanines are a large family of compounds that have been
extensively studied as sensitive materials for electrochemical sensors
due to their well-known electrocatalytic properties1. These metal
complexes act as mediators by lowering the overpotential of oxidation
or reduction of the target molecules or by increasing the intensity of the
observed peaks. Therefore, phthalocyanine sensors have been applied in
71
amperometric, voltammetric or potentiometric electrocatalytic

determination of several organic and inorganic compounds2,3.
Metallophthalocyanine complexes (MPcs) are the most widely studied

derivatives as electrochemical sensors2,5. In MPcs, the phthalocyanine
ring (in oxidation state −2) is coordinated with a range of transition
metal ions (in oxidation state +2). The number of works published using
MPcs as sensitive material is large and includes a variety of
phthalocyanine derivatives (e.g. central metal ions, substituents),
electrode designs, preparation techniques and target molecules2-6. Other
families of phthalocyanines such as the bisphthalocyanines (LnPc2) have
been less studied. LnPc2 are sandwich type compounds with free radical
character and ring oxidation state -1.57-9. This specific electronic
structure, characterized by the presence of an unpaired electron and by
the intramolecular interaction between the ligand systems, allows
sandwich complexes to find applications in many fields, including the
fabrication of sensors10-12. LnPc2 have demonstrated to be particularly
interesting sensing materials for voltammetric sensors due to their
remarkable electrochemical and electrocatalytic properties13,14. They
have been successfully used to detect phenolic compounds, a complex
group of substances of great interest due to their antioxidant activity15.
The control of the structure at the nanometric level (size, orientation,

alignment, thickness, etc.) is an important tool used to modulate the
sensor response. Nanostructured films prepared by Langmuir-Blodgett
(LB)6,16,17, Layer-by-Layer (LbL)18,19, self-assembling20 or electrodeposition
(EDP) techniques12 exhibit enhanced surface to volume ratios and well
controlled structures that facilitate the diffusion of ions inside the film.
For these reasons, nanostructured sensors show higher sensitivity, faster
kinetics and better reproducibility than non-nanostructured films16,21.
The sensing properties of bisphthalocyanines can be modulated by

varying the nature of the central rare earth ion and/or by introducing
substituents in the phthalocyanine rings4,22,23. Unlike the tetra-
substituted compounds, which are normally formed as a mixture of
isomers and used as such, the octa-substituted derivatives can be
72
synthesized isomerically pure, a feature which should encourage

molecular ordering in nanostructured films.
3.2.1- Chemicals
The bis[2,3,9,10,16,17,23,24-octachlorophthalocyaninate] lutetium (III)
complex (LuPc2Cl32), was synthesized according to a previously published
procedure24 and the molecular structure is shown in Figure 3.1. The
solvents (Merck) were of HPLC grade. The ultrapure water (18.2 MΩ·cm)
was acquired from a Milli-Q water purification system model Simplicity.
Catechol and other reactants were purchased from Sigma-Aldrich.
Figure 3.1. Structure of the LuPc2Cl32.
3.2.2- Langmuir, LB and LS films

The LuPc2Cl32 Langmuir, LS and LB films were fabricated using a Langmuir
trough KSV model 2000. The Langmuir films were produced by spreading
1000 L of 8.410-5 mol·L-1 solution of LuPc2Cl32 dissolved in THF onto
the water subphase. The Langmuir monolayers were characterized by
surface pressure versus mean molecular area (π–A) isotherms at 21oC
using the Wilhelmy method. The monolayer was symmetrically
compressed under a constant barrier speed at 10 mm·min-1.
The LB and the LS films were obtained by transferring the floating

Langmuir monolayers from the air/water interface to different solid
substrates depending on the characterization technique to be applied.
The surface pressure was kept constant at 30 mN·m-1. For LB films
upstroke and downstroke speeds of 0.5 to 3.0 mm·min-1 were used.
Under such conditions Y-type LB films with a transfer ratio close to unity
73
were obtained. LB and LS films containing up to 10 layers were

deposited onto quartz substrate for UV-Vis absorption spectroscopy
characterization. 20 LB layers and 10 LS layers were deposited onto ZnSe
for FTIR analysis in transmission mode and 10 LS layers were deposited
onto Ag mirror for the FTIR analysis in reflection mode. LB and LS
monolayers were deposited on Ag islands (evaporated film of Ag
containing 6 nm thickness) and glass for Surface-Enhanced Resonance
Raman Scattering (SERRS) and Resonance Raman Scattering (RRS)
measurements, respectively. For cyclic voltammetry and sensing studies
4 and 8 LS layers and 10 and 20 LB layers were deposited onto ITO
electrodes.
3.2.3- Film characterization

The UV-Vis absorption spectroscopy analysis was carried out using a
Varian spectrophotometer (model Cary 50) from 190 to 1000 nm. FTIR
were performed in a Bruker spectrometer model tensor 27 with spectral
resolution of 4 cm-1, 128 scans in the transmission mode. The reflection
spectra were recorded at an incident angle of 80o using an A118 Bruker
accessory. The Raman analysis was carried out using a micro-Raman
Renishaw spectrograph model in-Via equipped with a Leica microscope,
CCD detector, laser at 633 nm, and 1800 grooves·mm-1 gratings. In this
system a 50X objective allows collecting spectra with ca. 1 μm2 spatial
resolution.
The cyclic voltammetry (CV) was carried out using an EG&G PARC 263A
potentiostat/galvanostat (M270 Software) with a conventional three-
electrode cell. The reference electrode was Ag|AgCl/KCl sat. and the
counter electrode was a platinum plate. Cyclic voltammograms were
registered from -1.0 up to +1.0 V at a scan rate of 0.1 V·s-1 (except
otherwise indicated).
3.3.1- Langmuir films

The -A isotherms were recorded by spreading a THF solution of
LuPc2Cl32 onto an ultrapure water (MilliQ) subphase kept at a constant
74
temperature at 21oC. After solvent evaporation (20 minutes), the

phthalocyanine molecules were compressed at a constant rate of 10
mm·min-1 until the film pressure rose sharply, indicating that a
continuous surface film was formed (condensed phase). Reproducible -
A isotherms were recorded and some of them are given in Figure 3.2.a.
The extrapolated area (Aext in Figure 3.2.a) estimated by extrapolating
the condensed phase to =0 was 52Å2, smaller than the values observed
for unsubstituted LuPc216,25 and for LuPc2 substituted derivatives26-28. The
limiting area obtained here for LuPc2Cl32 suggests an edge-on
orientation; the face-on orientation (lying flat) would occupy an area of
approximately 160 and 170 Å2 29-31. The stacking of the LuPc2Cl32
molecules forming molecular aggregates is another possibility. For
instance, an aggregate with 3 molecules in a face-on orientation would
also lead to an extrapolated area of ca. 53 Å2. The monolayer showed a
collapse with a surface pressure of 40 mN·m-1, which is close to the
collapse observed for the unsubstituted and substituted LuPc225 and
indicates high molecular packing.
75
Figure 3.2. a) -A isotherms registered at 21 C. b) Stability test at 30 mN·m . c)

o -1
compression/expansion cycles for LuPc2Cl32.
Stability tests (i.e. displacement of the barriers to keep constant the

surface pressure in a value within the condensed phase of the film)
proved that the LuPc2Cl32 monolayer is very stable at the condensed
phase (Figure 3.2.b). A decrease of ca. 10% in the mean molecular area
was observed during the first 20 minutes, keeping constant after that.
Successive compression/expansion curves until the condensed phase
(surface pressure at 30 mN·m-1) revealed a hysteresis of the
thermodynamic process (Figure 3.2.c). It can be noted that the second
compression was shifted to lower areas however, for the first and the
second expansion processes the same mean molecular area was
reached. The latter is consistent with the formation of molecular
aggregates at the air/water interface during the first compression, which
is not reversible with expansion. Besides, the same area values at the
condensed phase found for both compressions indicate the Langmuir
film reaches the same structure, independent of the
compressing/expansion cycles.
3.3.2- LB and LS films. Growth monitored by UV-Vis absorption

The Langmuir films were transferred onto solid substrates at a pressure
of 30 mN·m-1 using both LS and LB techniques. The electronic absorption
spectra for the LuPc2Cl32 in THF solution, LB film (5 up to 10 layers) and
LS film (2 up to 10 layers) deposited onto a quartz substrates are given in
Figure 3.3. The UV-Vis spectrum of the LuPc2Cl32 THF solution showed
the characteristic features of macrocyclic compounds. The spectrum
presents strong B and Q bands that have been assigned to *
transitions. The Q band is centered at 697 nm and is considerably red-
shifted compared to Q band maximum (659 nm) observed in the
unsubstituted LuPc2 in chloroform solution32. This bathochromic shift is
usually observed when the aromatic ring is substituted with electron
withdrawing groups24,33. In addition, the substitution with eight acceptor
chlorine groups caused a strong decrease of the charge transfer bands
(associated with intramolecular transitions of the unpaired electron)
76
that appear at 450 nm and 900 nm in unsubstituted analogues, but

those were almost absent when strong withdrawing chlorine groups
were present. By registering spectra at increasing concentrations
(8.410-5, 6.7210-5, 4.210-5 and 1.6810-5 mol·L-1) a molar extinction
coefficient of 5.9104 M-1·cm-1 at 697 nm was obtained (spectra not
shown).
The UV-Vis spectra of the LB and LS films were similar to that of the THF
solution but a red shift of the Q band was observed in both cases. This
band appears centered at 711 nm for the LB film and at 709 nm for the
LS one. The shift observed in relation to maximum absorption of the
solution at 697 nm is characteristic of a face-to-tail stacking of the
chromophores (J aggregates)34, consistent with -A isotherm data. In
terms of film growth, the absorbance increased linearly with the number
of monolayers for both LB and LS films as shown in Figures 3.3.a and
3.3.b, confirming the good quality of the transferred films. Besides the
absorbance intensity for the Q band, comparison of 10 layers LB and LS
films indicated that a major quantity of material was transferred to the
substrate in the LS film for each layer, in average. It is important to note
that layers were deposited only onto one side of the substrate for LS
films while layers were deposited onto both sides of the substrate for LB
films. Finally, focusing on Q band, the greater relative intensity of the
band at higher wavelength for both films indicated the monomers were
dominant in relation to dimers or higher order of aggregates of
LuPc2Cl32.
77
Figure 3.3. UV-Vis absorption spectra for a) LB film and b) LS film of LuPc 2Cl32up
to 10 layers deposited onto quartz plates. The dashed red line in 3.a
corresponds to the THF solution spectrum of LuPc 2Cl32 (out of scale).
3.3.3- LB an LS films. Molecular organization determined by

FTIR
The FTIR is a powerful tool to determine possible molecular orientation
in thin films35. Therefore, FTIR spectra of LB (transmission mode) and LS
films (transmission and reflection modes) films of LuPc2Cl32 are
compared Figure 3.4. Table 3.1 shows a list of the active vibrations and
their assignment in the LB and LS films36. An important point to be
noticed here is that according to the surface selection rules37 the
fundamental vibrations with transition dipole parallel to the substrate
surface have maximum intensity for the FTIR spectra obtained in
transmission mode (incident electric field parallel to the substrate
surface). In reflection mode (incident electric field polarized
perpendicularly to the substrate surface), the intensities of the
fundamental vibrations with dynamic dipole perpendicular to the
surface have maximum intensity37.
Table 3.1. Infrared active vibrations and their assignment in the LB and LS films.
Wavenumber (cm-1) Assignment

759 C-Cl stretching
932 benzene ring
1091 phthalocyanine ring breathing
1135 pyrrole ring breathing
1205 isoindole stretching
1653 benzene stretching
1733 C=O stretching (from acetate)
78
In Figures 3.4.a and 3.4.b, the FTIR spectra for the LS film in both
transmission and reflection modes presented a similar profile. This
indicates a non-preferential molecular arrangement or an orientation of
the phthalocyanine ring forming ca. 45o in relation to the substrate for
LS films38. A similar profile was also observed comparing the spectra in
Figures 3.4.a and 3.4.c (i.e. the FTIR spectra for LS and LB films in
transmission mode) indicating that both LS and LB presented the
LuPc2Cl32 structured in a similar way. Besides, all spectra were similar to
that observed to LuPc2Cl32 in KBr pellet given by Gobernado-Mitre et
al.36. Therefore, because the FTIR spectrum of the powder is
characteristic of randomly structured system, it can be concluded that,
despite LB and LS grow in a well stratified way, they presented a non-
preferential molecular arrangement.
Figure 3.4. FTIR spectra in transmission mode for both a) 10-layer LS film and c)
20-layer LB film, deposited on ZnSe and in reflection mode for b) 10-layer LS
deposited on Ag mirror.
3.3.4- Surface-Enhanced Resonance Raman Scattering (SERRS)

The SERRS spectra of both LB and LS monolayers onto Ag island films (6
nm of Ag evaporated onto glass slides, leading to Ag nanoparticles) are
shown in Figure 3.5.a. It was observed that the SERRS spectra for both
LB and LS monolayers presented the same profile. This SERRS profile was
the same shown by the LuPc2Cl32 powder (result not shown) and similar
79
to that presented in Ref. 36. The simply enhancement of the RRS spectra
without any change in the spectra profile, as the relative intensity of the
bands, is a common observation for SERRS of phthalocyanines39. This
absence of changes in the SERRS spectra indicated that there was no
chemical interaction between molecule-metal (LuPc2Cl32-Ag) and both LB
and LS monolayers were physisorbed on the Ag islands.
Figure 3.5. a) SERRS spectra for both LB and LS monolayers of LuPc 2Cl32 onto Ag
islands films (Ag nanoparticles). b) RRS and SERRS spectra for LB monolayers on
glass and Ag islands, respectively.
The enhancement factors (EF) can be estimated for both LB and LS films
considering the intensity ratio SERRS/RRS of the band at 1525 cm-1. For
instance, for the LB film (Figure 3.5.b) the intensity of the SERRS band at
1525 cm-1 was ca. 32,500 counts and in the RRS the band intensity was
ca. 3,500 counts (the intensity is the height from the top to the bottom
of the band). Considering the intensity ratio SERRS/RRS for this band,
and that the RRS spectrum was collected with a laser power 200 times
higher than the SERRS spectrum, the enhancement factor was estimated
to be ca. 2103. A similar EF was observed for the LS film. These EF
observed are in agreement with those found to phthalocyanine
molecules40,41 and in agreement with the model considering the
electromagnetic mechanism35,42. Basically, according to the EM
mechanism, the enhancement of the Raman signal was achieved by the
80
localized surface plasmon resonances (LSPR), which were sustained by

the Ag nanoparticles and led to the enhacement of the electric field
surrounding the Ag nanoparticles35,42. Therefore, the laser line (633 nm)
must be in resonance with the Ag plasmon absorption (necessary
condition) to achieve the surface-enhanced Raman scattering (SERS). In
our case, because the 633 nm laser line was also in resonance with the
LuPc2Cl32 absorption (Figure 3.3.a), the resonance Raman scattering
(RRS) was achieved. Therefore, when the LuPc2Cl32 monolayer was
deposited onto the Ag island films and excited with the 633nm laser line,
a double resonance was established and the SERRS phenomenon was
achieved43.
3.3.5- Electrochemical properties

In order to evaluate the electrochemical behavior of the LuPc2Cl32
sensors, the electrodes were immersed in a KCl solution and cyclic
voltammograms were registered from –1.0 to +1.0 V at a scan rate of 0.1
V·s-1. Figure 3.6 shows the cyclic voltammograms of LB and LS films of
LuPc2Cl32 deposited onto ITO glass in contact with 0.1 mol·L-1 KCl
aqueous solution. The most remarkable aspect of the cyclic
voltammograms was the absence of the peaks attributed to the quasi-
reversible processes corresponding to one electron oxidation
Ln(III)Pc2/Ln(III)Pc2+ and the one electron reduction Ln(III)Pc2/Ln(III)Pc2-
of the phthalocyanine ring, which are usually observed in unsubstituted
LnPc211,22,28,44. This absence can be explained taking into account the
computed ionization potentials and the electro-affinity published for
asymmetrically substituted chlorine bisphthalocyanines45. According to
these data, the halogen electro-attracting inductive effect tends to
stabilize the phthalocyanine frontier orbital. The presence of electron
acceptor chlorine groups hinders the oxidation and reduction of the
phthalocyanine ring. For this reason, the peaks corresponding to these
processes were not observed in the studied range.
81
Figure 3.6. CV of LuPc2Cl32 LB (10 monolayers) and LS (8 monolayers) films

-1
immersed in 0.1 mol·L . Blue line corresponds to LB films and red line
-1
corresponds to LS films. Scan rate, 0.1 V·s .
Voltammograms also showed a cathodic wave at -0.90 V that was
caused by the decomposition of water at the electrode surface. This
peak was also observed at -1.1 V when using neat ITO electrodes. The
shift of the potential at which decomposition of water occurs, was due
to the electrocatalytic effect of the phthalocyanine layer that facilitates
the reduction process. The intensity of the cathodic wave was
dependent on the pH of the solution: it decreased when the pH
increased (result not shown), confirming the assignment of this peak.
The results obtained from LB and LS were similar and only a difference in
the decomposition of water (favored at LS electrodes) was observed.
3.3.6- Electrochemical detection of catechol

As stated before, unsubstituted or octa-terbutyl bisphthalocyanines
immobilized on inert electrode surfaces, exhibit electrocatalytic activity
towards phenolic compounds14,15,24. The presence of bisphthalocyanine
resulted in a decrease in the overpotential of oxidation or reduction of
the target molecules and/or an increase in the intensity of the peaks
observed.
In order to evaluate the electrocatalytic activity of the LuPc2Cl32 films

towards phenols, the electrochemical response of 10-3 mol·L-1 catechol
in 0.1 mol·L-1 KCl at LuPc2Cl32 modified ITO electrodes was tested at a
scan rate of 0.1 V·s-1. For comparison purposes, the cyclic
voltammogram of catechol for a bare ITO electrode was also recorded
82
(Figure 3.7). In good acordance with the literature, when a bare ITO glass
was used as electrode, the voltammogram showed an anodic peak (at
0.55 V) in the positive-going scan and a cathodic counterpart peak (at
0.22 V) in the negative-going scan. These peaks correspond to the
transformation of catechol to o-benzoquinone and vice-versa in a quasi-
reversible two-electron process46-48. Under these conditions, ΔE is 0.33 V
and peak current ratio (Ic/Ia) is close to unity.
Voltammograms obtained using electrodes modified with LuPc2Cl32 films

immersed in a catechol solution, differed significantly from
voltammograms recorded using a bare ITO glass electrode. For instance,
when a cast film was exposed to catechol (Figure 3.7), voltammograms
showed the expected redox pair associated with the oxidation/reduction
of catechol48. The cathodic peak appeared at 0.22V (the same potential
observed when using a bare ITO glass), whereas the anodic peak was
shifted to lower potentials and appeared at 0.4 V (0.55 V at ITO
electrode). The peak potential difference (ΔE) between anodic and
cathodic peak (ΔE=0.18 V) was smaller than the observed for bare ITO
glass (ΔE=0.33 V), indicating that the reversibility of electrochemical
reaction was improved in the presence of LuPc2Cl32. This electrocatalytic
activity has been already observed in electrodes modified with
unsubstituted LuPc2 where the oxidation of catechol occurs at 0.47 V
with ΔE=0.30 V16,32. In this case, when the phthalocyanine ring was
substituted with chlorine atoms, the shift to lower potentials was more
pronounced.
It is worth to remark that the anodic peak at 0.4 V was accompanied by

a shoulder at 0.25 V. This splitting might be due to the existence of two
types of binding sites for immobilization, giving rise to active sites with
distinct electrocatalytic efficiencies. Further proofs of the existence of
two types of active sites will be shown in the dynamic study of the
electrodes at a different scan rates. In addition, a cathodic wave at -0.5
V was also observed as a result of the irreversible reduction of catechol
catalyzed by the phthalocyanine.
83
-3 -1
Figure 3.7. Cyclic voltammogram of 10 mol·L catechol in the presence of 0.1
-1
mol·L KCl at a bare ITO electrode (red line) and a cast film of LuPc2Cl32 (blue
-1
line). Scan rate, 0.1 V·s .
It is also important to notice that the redox peak currents of catechol

increased about a 25% with respect to the intensity observed for a bare
ITO glass (the increase in intensity observed in unsubstituted LuPc2 is
tenfold the observed when using bare ITO glass). In summary, the
excellent electrocatalytic activity of LuPc2Cl32 is responsible for the
significantly improved electrochemical behavior of catechol at the
LuPc2Cl32 cast film.
It is well known that the structure of the sensing layer plays a key role in
the electrochemical behavior of modified electrodes. The
voltammograms obtained using LuPc2Cl32 nanostructured films
immersed in a catechol solution differed significantly from
voltammograms recorded using a bare ITO glass or an ITO electrode
modified with a cast film. This is illustrated in Figure 3.8 where the
electrochemical responses of LB and LS films immersed in a solution of
10-3 mol·L-1 catechol (in 0.1 mol·L-1 KCl) are shown.
84
-3 -1
Figure 3.8. Cyclic voltammograms of 10 mol·L catechol in the presence of 0.1
-1
mol·L KCl at 20 monolayers LB films (green line), 10 monolayers LB (red line)
-1
and 8 monolayers LS (blue line) of LuPc2Cl32. Scan rate, 0.1·V·s .
When LuPc2Cl32 LB or LS electrodes were immersed in catechol, a set of

anodic (positive) and cathodic (negative) peaks were observed. For more
clarity, peaks have been numbered. The oxidation of the catechol to a
quinone gave rise to a broad anodic peak on the forward scan of the
cyclic voltammograms (peak I), while reduction of the quinone back to
the original polyphenol produced the corresponding cathodic peak on
the reverse scan (peak III). Due to the electrocatalytic activity of the
LuPc2Cl32, peak I was displaced to 0.27 V for both LB and LS (0.55 V when
using a bare ITO glass electrode and 0.4 V when using a cast film). In
turn, peak III was also shifted to lower potentials (at 0.1 V for LB, -0.05 V
for LS and 0.2V for cast film). The anodic peak I associated with the
oxidation of catechol was quite broad and by similarity with the cast
film, it can be assumed that this broad peak was the wave resulting from
the overlapping of redox processes at two different active sites.
An important difference between LB (or LS) and cast films is that in

nanostructured films a new intense anodic wave was observed at 0.75 V
(peak II). According to the literature, this new redox process can be
associated with the polymerization of the previously oxidized cathecol44-
49
. This peak was not observed when using a LuPc2Cl32 cast film,
indicating that the structure of the surface and the homogeneous
molecular packaging of the nanostructured films facilitated the second
85
oxidation. The polymeric form was reduced at potentials close to 0.2 V.

That is, in the cathodic sweep, the peak observed at 0.2 V (peak III)
corresponded to the overlapping of the reduction of both the quinoid
form and polymeric form of catechol, justifying thus the increase in
intensity and in broadness with respect to the cast film. The cathodic
wave at -0.5 V was also observed (peak IV).
The enhanced electrocatalytic effect observed in LB and LS films can be

attributed to the LuPc2Cl32 molecular arrangement/packing. The
homogeneous layered structure provided well-defined surfaces with an
increased number of active sites. The small structural differences
between LB and LS films described in sections 3.3.1 to 3.3.4 justify some
discrepancies in their electrochemical behavior. For instance, the
oxidation and the polymerization of catechol (peaks I and II) were
favored at the surface of LS films. An important difference between LB
and LS films is their stability. It is well known that in chemically modified
electrodes, the first cycles differ from subsequent ones. For this reason,
before using the electrodes for sensing purposes, it is necessary to run
several stabilization cycles. In this case it was observed that LB films
were stabilized after three cycles, whereas, LS films were not stabilized
until fifteen cycles were run (all the voltammograms shown in the
figures correspond to stabilized signals). After stabilization, SD values
(calculated from the peak I intensity values) were lower than 5% for
both LB and LS films.
The thickness also influenced the electrochemical response of LB or LS

films exposed to catechol (Figure 3.8). In both types of films, peaks
associated with the polymerization of the catechol (peaks II and III)
increased in intensity when increasing the number of monolayers. This
effect can be explained by assuming that the inner layers facilitate the
transfer of electrons to the ITO. In contrast, by increasing the number of
monolayers the intensity of the peaks I and IV, associated exclusively
with the oxidation/reduction of catechol, did not change. This means
that these redox processes occur exclusively at the active sites located at
the electrode surface.
86
3.3.7- Dynamic behavior

The dynamic character of the electrode was examined by registering
voltammograms at different scan rates. As observed in Figure 3.9, the
intensity of the voltammograms increased with the scan rate. At high
scan rates, the splitting of peak I became more evident, confirming the
existence of two active sites at the film surface.
The inset of Figure 3.9 shows the linear correlation between the peak
current (peak I) and the scan rate in the range 0.05-0.5 V·s-1 (Table 3.2).
As shown in Figure 3.9, the intensity of peak I increased linearly with the
scan rate, indicating the dominance of the surface confined processes.
Under these conditions the surface coverage ( can be calculated using
the Laviron equation (Equation 3.1):
I=nF2A/4RT (Equation 3.1)
Where n is the number of electrons, F the Faraday constant (C·mol-1), 

the scan rate (V·s-1), A the area of the sensor (cm2),  the surface
coverage (mol·cm-2), R the gas constant (cm3·atm·K−1·mol−1) and T the
temperature (K). The surface coverage of the LuPc2Cl32 films calculated
using Equation 3.1 was =7.7744910-11 mol·cm-2 for the LB film and
=1.2755210-10 mol·cm-2 for the LS film. The inherent irreproducibility
of cast films did not allow calculating the surface coverage accurately.
These values were higher than those obtained using a bare ITO glass
electrode (2.3710-11 mol·cm-2) indicating that the presence of the
nanostructured layer of LuPc2Cl32, increased the number of active sites,
thus facilitating the oxidation of the catechol.
87
Figure 3.9. Cyclic voltammogram of a LS sensor (8 layers) registered at different

scan rates from 0.05 to 0.5 V·s . The inset shows the plot of Ia vs.  (calculated
-1
-3 -1 -1
for peak I). Electrolyte solution was 10 mol·L cathecol (in KCl 0.1 mol·L ).
2
Table 3.2. Slope (m), coefficient of correlation (R ) and surface coverage ()
obtained by representing the intensity of the anodic peak I and the scan rate
().
Electrode m (slope) R2 (mol·cm-2)
LuPc2Cl32/20LB 710-5 0.995 7.774510-11
LuPc2Cl32/8LS 110-4 0.996 1.275510-10
3.3.8- Limits of detection

The intensity of peak I was linearly dependent on the concentration of
catechol in the range from 610-5 to 510-4 mol·L-1 as shown in Figure
3.10. The corresponding detection limits were calculated according to
the 3sd/m criterion, where m is the slope of the calibration graph, and sd
was estimated as the standard deviation (n=7) of the voltammetric
signal of the electrode at the concentration level corresponding to the
lowest concentration of the calibration plot. The detection limits
calculated were 7.4910-5 mol·L-1 for LB and 8.4210-5 mol·L-1 for LS.
These results demonstrate that the LuPc2Cl32 nanostructured films can
be used to quantify the presence of catechol in the range usually
present in foods.
88
Figure 3.10. Calibration curve obtained from the LB film sensor (20
monolayers).
3.4 CONCLUSIONS
LB and LS films of bis[2,3,9,10,16,17,23,24-octachlorophthalocyaninate]
lutetium(III) complex were successfully prepared forming well stratified
layers. The molecular packing induced by the presence of the
substituents differed from that showed by other bisphthalocyanine
compounds and strongly affected the electrochemical properties of the
LB and LS films when compared to LuPc2Cl32 cast film and bare ITO glass
electrode. The electrocatalytic activity of the LuPc2Cl32 and the abundant
binding sites for the immobilization provided by the nanostructured
films are responsible for the significantly improved electrochemical
behavior of catechol at the LuPc2Cl32 films.
The small structural difference between LB and LS films justifies some

divergence in their sensing properties. For instance, the oxidation and
the polymerization of catechol are favored at the surface of LS films,
whereas LB films are more stable.
In summary, the electrochemical results indicate that the LuPc2Cl32 LB

and LS films modified ITO glass electrodes ca be used as catechol sensing
units with good stability and detection limit around 10-5 mol·L-1.
89
References
1. Jiang, J. Functional phthalocyanine molecular materials, Series:
Structure and Bonding. 2010, Ed. Springer. Berlin (Germany)
2. Lever, A.B.P. J. Porphyrins Phthalocyanines. 1999, 3, 488-499.
3. Zagal, H., Griveau, S., Silva, J.F., Nyokong, T., Bedoui, F. Coord. Chem.
Rev. 2010, 254, 2755-2791.
90
4. Rodríguez-Méndez, M.L. Sensing properties of Phthalocyanines in

Encyclopedia of Sensors, 2006, 111-134. Ed. America Scientific
Publishers. California (USA)
5. Bedoui, F., Griveau, S., Nyokong, T., Appleby, A.J., Caro, C.A., M.
Gulppi, M., Ochoa, G., Zagal, J.H. Phys. Chem. Chem. Phys. 2007,9, 3383-
3396.
6. Valli, L. Adv. Col. Interf. Sci. 2005,116, 13-44.
7. Weiss, R., Fischer, J. Lanthanide Phthalocyanine Complexes in

Porphyrin Handbook. 2010, 171-246. Ed. Academic Press, New York
(USA).
8. Komorsky-Lovri´c, S., Quentel, F., L’Her, M., Scholz, F. J. Solid State

Electrochem. 2008, 12, 165-169.
9. Bouvet, M. Radical phthalocyanines and intrinsic semiconduction, in

Porphyrin Handbook, 2010, 37-104. Ed. Academic Press, New York
(USA).
10. Rodríguez-Méndez, M.L., Gay, M., de Saja, J.A. J. Porphyrins

Phthalocyanines. 2009, 13, 1159-1167.
11. Jiang, J.Z., Liu, R.C.W., Mak, T.C.W., Chan, T.W.D., Ng, D.K.P.
Polyhedron. 1997, 16, 515-520.
12. Gay, M., Muñoz, R., de Saja, J.A., Rodriguez-Mendez, M.L.

13. Yilmaz, I., Nakanishi, T., Gurek, A.G., Kadish, K.M. J. Porphyrins.
14. Gorbunova, Y., Rodríguez-Méndez, M.L., Kalashnikova, I.P., Tomilova,

L., de Saja, J.A. Langmuir. 2001, 17, 5004-5010.
15. Njanja, E., Nassi, A., Ngameni, E., Elleouet, C., Quentel, F., L’Her, M.
Electrochem. Commun. 2007, 9, 1695-1699.
91
16. Apetrei, C., Nieto, M., Rodríguez-Méndez, M.L., de Saja, J.A. J.

Porphyrins Phthalocyanines. 2011, 15, 1-10.
17. Aoki, P.H.B., Caetano, W., Volpati, D., Riul, A.,Constantino, C.J.L. J.
Nanosci. Nanotechnol. 2008, 8, 4341-4318.
18. Apetrei, C., Apetrei, I., de Saja, J.A., Rodriguez-Méndez, M.L. Sensors.
2011, 11, 1328-1344.
19. Centurion, L.M.P.C., Moreira, W., Zuccolotto, V. J. Nanosci.

Nanotechnol. 2012,12, 2399-2405.
20. Khaderbad, M.A., Tjoa, V., Mathews, N., Mahdu,S., Ravikan, M.,
Rao,V.R. Nanosci. Nantechnol. Let. 2012, 4, 743-746.
21. Narducci, D. Sci. of Advanced Mat. 2011, 3, 426-435.
22. Arrieta, A. Parra, V., Rodríguez-Méndez, M.L., de Saja, J.A. Sensors

Actuat. B. 2003, 95, 357-365.
23. Fernandes, E.G.R., Brazaca, L.C., Rodríguez-Méndez, M.L., de Saja,

J.A., Zucolotto, V. Biosens. Bioelectron. 2011, 26, 4715-4719.
24. Pushkarev, V.E., Shulishov, E.V., Tomilov, Y.V., Tomilova, L.G.

Tetrahedron Letters. 2007, 48, 5269-5273.
25. Alessio, P., Pavinatto, F.J., Oliveira, O.N. Jr., de Saja Saez, J.A.,
Constantino, C.J.L., Rodriguez-Mendez, M.L. Analyst. 2010, 135, 2591-
2599.
26. Jones, R., Hunter, R.A., Davidson, K. Thin Solid Films. 1994, 250, 249-
257.
27. Selektor, S.L., Shokurov, A.V., Raitman, O.A., Sheinina, L.S., Arslanov,
V., Birin, K.P., Gorbunova, Y.G., Tsivadze, A.Y. Colloid Journal. 2012, 74,
334-345.
28. Rodríguez-Méndez, M.L., Aroca, R., de Saja, J.A. Chem. Mater. 1993,
5, 933-937.
92
29. Gaffo, L., Constantino, C.J.L., Moreira, W.C., Aroca, R.F., Oliveira,
O.N.Jr. Spectrochim. Acta-Part A. 2004, 60, 321-327.
30. Clavijo, R.E., Battisti, D., Aroca, R., Kovacs, G.J., Jennings, C.A.
Langmuir. 1992, 8, 113-117.
31. Maitrot, M., Guillaud, G., Boudjema, B., Andre, J.J., Strzelecka, H.,
Simon, J., Even, R. Chem. Phys. Lett. 1987, 133, 59-62.
32. Gay Martin, M., Rodrıguez-Mendez, M.L., de Saja, J.A. Langmuir.

2010, 26, 19217-19224.
33. Rousseau, R., Aroca, R., Rodríguez-Méndez, M.L. J. Molecular

Structure. 1995, 356, 49-62.
34. Rousseau, E., Koets, M.M., Van der Auweraer, M., De Schryver, F.C.
Photochem. Photobiol. Sci. 2002, 1, 395-406.
35. Aroca, R. Surface-Enhanced Vibrational Spectroscopy. 2006, Ed. John

Wiley and Sons. Hoboken, (USA).
36. Gobernado-Mitre, I., Klassen, B., Aroca, R., de Saja, J.A. J. Raman
Spectroscopy. 1993, 24, 903-908.
37. Debe, M.K. Prog. Surf. Sci. 1987, 24, 1-4.
38. Alessio, P., de Oliveira, R.F., Aoki, P.H.B., Pereira, J.D.A.S., Braunger,
M.L., Furini, L.N., Vieira, M., Teixeira, S.R., Job, A.E., Saenz, C.A.T., Alves,
N., Olivati, C.A., Constantino, C.J.L. J. Nanosci. Nanotechnol. 2012, 12,
7010-7020.
39. Aroca, R., Kovacs, G.J., Vibrational Spectra and Structure, 1991, Ed.
Elsevier, Amsterdam (Netherlands).
40. Alessio, P., Rodrıguez-Mendez, M.L., de Saja Saez, J.A., Constantino,

C.J.L. Phys. Chem. Chem. Phys. 2010, 12, 3972-3983.
41. Alessio, P., de Saja Saez, J.A., Aroca, R.F., Constantino, C.J.L. Applied
Spectroscopy. 2011, 65, 152-158.
93
42. Le Ru, E.C., Etchegoin, P.G. Principles of Surface Enhanced Raman

Spectroscopy (and Related Plasmonic Effects). 2009, Ed. Elsevier,
Amsterdam (Netherlands).
43. Alessio, P., Constantino, C.J.L., Aroca, R.F., Oliveira, O.N.Jr. J. Chil.
Chem. Soc. 2010, 55, 469-478.
44. Yilmaz, I., Nakanishi, T., Gurek, A., Kadish, K.M. J. Porphyrins
45. Kahlal, S., Mentec, A., Pondaven, A., L’Her, M., Saillard, J.Y. New J.
Chem. 2009, 33, 574-582.
46. Nematollahi, D., Alimoradi, M., Husainc, S.W. Electroanalysis. 2004,

16, 1359-1365.
47. Wang, G., He, X., Zhou, F., Li, Z., Fang, B., Zhang, X., Wang, L. Food
Chem. 2012, 135, 446-451.
48. Makhotkina, O., Kilmartin, P.A. Anal. Chim. Acta. 2010, 668, 155-165.
49. Huang, W., Luo, S., Zhou, D., Zhang, S., Wu, K.B. Nanosci.
Nanotechnol. Lett. 2013, 5, 367-371.
94
CHAPTER 4
Synergistic Electrocatalytic Effect of
Nanostructured Mixed Films Formed by
Functionalised Gold Nanoparticles and
Bisphthalocyanines
Once demonstrated the electrocatalytic effect that Langmuir-Blodgett-

films of phthalocyanines and gold nanoparticles present towards the
detection of compounds with interest in the enology field, the goal of
this chapter was to analyze the effect of co-depositing two
electrocatalytic materials.
A new methodology has been developed to co-deposit LB films of

phthalocyanines with gold nanoparticles. One of the achievement of this
work was to use LB technique to modify the substrate (ITO) with polar
gold nanoparticles modified. The method consisted on the insertion of
the water soluble gold nanoparticles [(11-mercaptoundecyl)
tetra(ethylene glycol)] (SAuNP) underneath a floating Langmuir film
formed by a phthalocyanine [lutetium (III) bisphthalocyanine] (LuPc2)
and an amphiphile [dimethyl dioctadecyl ammonium bromide] (DODAB)
required to decrease the rigidity of the film. The insertion of the SAuNPs
into the nanostructured LuPc2:DODAB floating film to obtain
LuPc2:DODAB/SAuNP films will be evaluated using spectroscopic and
microscopic techniques. The sensing capabilities of films containing
different proportions of LuPc2:DODAB and increasing amounts of
SAuNPs have been studied towards hydroquinone, an antioxidant
belonging to the phenols family commonly found in beverages such as
teas and wines, using cyclic voltammetry (CV). The synergistic effect
caused by the interaction of two electrocatalytic materials: SAuNPs and
LuPc2 will be discussed. This work was carried out in collaboration with
researchers from UNESP (Brazil) and has been published in Analytica
Chimica Acta.
4.1 INTRODUCTION
Phenolic compounds are a large family of molecules which are widely
distributed in plant-derived foods. They have attracted great interest
due to their potential health benefits. Polyphenolic compounds also
contribute to the organoleptic characteristics of foods and affect their
antioxidant capacity1. The determination of phenols can be carried out
using traditional techniques including spectroscopy, gas or liquid
chromatography and electrochemical methods. Oxidation of phenols is
97
feasible at inert electrodes (e.g. metallic or glassy carbon electrodes)2

and at electrodes chemically modified with a variety of materials and
biomaterials3. The modification of the electrode surface with
nanomaterials has demonstrated to be a good strategy to prepare
sensing devices4,5. For instance, electrodes modified with gold
nanoparticles (AuNPs) show unique catalytic properties that depend on
the AuNP size and on the nature of the protecting groups. In addition,
AuNPs provide a large surface area for the detection of analytes, leading
to electrochemical sensors with lower detection limits6-9.
On the other hand, metallophthalocyanines (MPc) and their sandwich

type lanthanide derivatives (LnPc2) have demonstrated to behave as
excellent sensing materials for the detection of a variety of compounds
due to their well-known electrocatalytic properties, fast electron
transfer and electron mediator capabilities10-18.
The design of hybrid systems combining different electrocatalytic

materials is very attractive for the development of electrochemical
sensors. Recently, AuNPs and phthalocyanines have been successfully
combined at a sensor surface using spin coating19-20. However, the
applications of these composite AuNP/MPc systems require the
development of methods allowing both classes of compounds to be co-
immobilized in structures with controlled distribution of the two
components21,22.
The LB technique can be an alternative to produce nanostructured

multilayers where two or more compounds are co-deposited in mixed
films23. The materials used to form the films must fulfil the particular
requirements of this technique. For instance, in classical LB deposition,
the film-forming materials must be amphiphilic molecules (insoluble in
water but soluble in organic solvents). This means that the AuNPs must
be capped (e.g. with thiolate groups) which facilitate the adequate
dispersion onto the water subphase24,25. In turn, to obtain
phthalocyanine multilayer assemblies, it is often necessary to mix the
phthalocyanines with amphiphiles such as fatty acids or phospholipids
98
that induce flexibility in the monolayers facilitating the transfer to solid

substrates15,18.
4.2.1- Langmuir monolayers and Langmuir-Blodgett films

The lutetium (III) bisphthalocyaninate (LuPc2) was synthesized and
purified in the neutral radical state using a previously published
method27. (11-mercaptoundecyl)tetra(ethylene glycol) functionalized
gold nanoparticles (SAuNP), dimethyldioctadecylammonium bromide
(DODAB) and all the chemicals and solvents were purchased from Sigma-
Aldrich and Fluka.
Buffer phosphate 0.01 mol·L-1 (pH 7.0) was prepared mixing sodium
phosphate monobasic (purity>99%) and sodium phosphate dibasic
(purity 99%) in ultrapure water.
LB films were prepared in a KSV 2000 System Langmuir-Blodgett trough

equipped with a Wilhelmy plate to measure the surface pressure versus
mean molecular area (-A) isotherms.
Films containing LuPc2 and DODAB (LuPc2:DODAB) were prepared by

spreading a chloroform solution of the corresponding mixture (10:1),
(1:1) or (1:10) (1·10-5 mol·L-1) onto a water subphase (ultrapure water-
Millipore MilliQ kept at 20oC). Then, the barriers were compressed at a
speed of 10 mm·min-1 to register the -A isotherms. At a surface
pressure of 35 mN·m-1, 20 monolayers were transferred to the solid
substrates by Z type deposition with a transfer ratio close to 1.
Films containing SAuNP, DODAB and LuPc2 (LuPc2:DODAB/SAuNP), were

prepared from chloroform mixtures of LuPc2:DODAB (10:1), (1:1) and
(1:10) that were spread onto the water subphase. After evaporation of
the solvent, 100 or 500 L of an aqueous solution of SAuNP (0.02% w/v)
were injected drop by drop underneath the air-water interface (before
compressing the film) and adsorbed inside the floating monolayer.
Floating layers of LuPc2:DODAB/SAuNP were compressed and
99
transferred to solid substrates using the same conditions mentioned in

the previous paragraph. 20 monolayers LB films were built by Z type
deposition with a transfer ratio close to 1. A scheme of the preparation
method is presented in Figure 4.1. The complete list of the sensors
prepared is collected in Table 4.1
Figure 4.1. Method to prepare the LuPc 2:DODAB/SAuNP films: DODAB, LuPc2,
SAuNP.
Table 4.1. List of the sensors prepared.
Proportion Vol. SAuNPs

Film
LuPc2:DODAB added (µL)
LuPc2:DODAB (10:1) 10:1 0
LuPc2:DODAB/SAuNP (10:1/100) 10:1 100
LuPc2:DODAB/SAuNP (10:1/500) 10:1 500
LuPc2:DODAB (1:1) 1:1 0
LuPc2:DODAB/SAuNP (1:1/100) 1:1 100
LuPc2:DODAB/SAuNP (1:1/500) 1:1 500
LuPc2:DODAB (1:10) 1:10 0
LuPc2:DODAB/SAuNP (1:10/100) 1:10 100
LuPc2:DODAB/SAuNP (1:10/500) 1:10 500
4.2.2- Films characterization

Langmuir films were analyzed by Brewster Angle Microscopy (BAM)
using a KSV MicroBAM. UV-Visible absorption measurements were
conducted using a Shimadzu spectrophotometer UV-1603. TEM images
were recorded using a JEM –FS2200 HRP instrument.
100
Raman analysis was performed in a Raman spectrometer UV-HR Lab

Ram from Horiba-Jobin Ybon, equipped with a LN2-cooled charge-
coupled device (CCD) detector. The excitation was carried out with
different lasers, including a Nd-YAG laser (emitting at 532 nm) and a He-
Ne laser (emitting at 633 nm), with a spectral resolution of 1 cm-1. The
sample was excited through the 50X optical objective of a Leica
microscope. The scattered light was collected through the same optical
objective (back-scattering configuration) with ca. 1 μm spatial
101
resolution. The excitation power on the sample surface was kept below
100 kW·cm-2.
The electrochemistry was carried out in an EG&G PARC 273

potentiostat/galvanostat using a conventional three-electrode cell. The
LB films were used as working electrodes. The reference electrode was
Ag|AgCl/KCl 3M and the counter electrode was a platinum plate.
4.3.1- Langmuir monolayers

The -A isotherms of mixtures of LuPc2:DODAB (proportions 10:1, 1:1
and 1:10) were registered by spreading the corresponding chloroform
mixture onto the water subphase and compressing the barriers at a
speed of 10 mm·min-1 while registering the variations of the surface
pressure (Figure 4.2.a). When the proportion of LuPc2 was high (10:1),
the isotherms were compatible with a total miscibility of the two
components. The calculated limiting area of 90 Å2 was in agreement
with the limiting area previously reported for LuPc2 and corresponds to
an edge-on orientation with the Pc ring perpendicular to the water
subphase28-30. When DODAB was the majoritarian component (1:10), the
-A isotherms showed a classical pattern of immiscibility consisting on
two regions of different slope. At low pressures, the slope of the curve
was 127 Å2 coinciding with the limiting area registered for neat
DODAB31, whereas in the second steep rise of the surface pressure, the
limiting area calculated was 90 Å2 that coincided with the value of neat
LuPc2. The -A isotherm of the equimolar mixture LuPc2:DODAB (1:1),
was consistent with a miscible system and showed a large area per
molecule of 240 Å2. This expansion of the monolayer could be due to
repulsive interactions between LuPc2 and DODAB or change of molecular
accommodation at the surface32.
102
Figure 4.2. Surface pressure/Area per molecule isotherms of a)

LuPc2:DODAB(10:1) (green line), LuPc2:DODAB(1:1) (red line) and LuPc2:DODAB
(1:10) (blue line). b) LuPc2:DODAB (1:1) (red line); LuPc2:DODAB/SAuNP
(1:1/100) (blue line) and LuPc2:DODAB/SAuNP (1:1/500) (green line).
SAuNPs were inserted in the LuPc2:DODAB floating films by injecting the

nanoparticles underneath the air-water interface. The insertion of the
gold nanoparticles in the floating films was confirmed by the expansion
of the area per molecule that increased with the amount of SAuNPs
injected under the surface (Figure 4.2.b). The expansion of the
monolayer was clear in LuPc2:DODAB mixtures 10:1 and 1:1. In contrast,
in 1:10 films (low concentration of LuPc2), the insertion of SAuNP was
almost negligible suggesting that the insertion of the SAuNP was mainly
due to the conjugation of SAuNPs to the phthalocyanine molecules. It is
also important to indicate that injecting amounts of SAuNPs higher than
500 l, caused irreproducibility, probably due to the occurrence of phase
segregation.
The quality of the floating films was further analyzed using Brewster
Angle Microscopy (BAM). In the absence of SAuNP, Langmuir films
showed highly homogeneous surfaces. Films containing SAuNP showed a
granular structure as a result of the insertion of the nanoparticles in the
floating film. This is illustrated in Figure 4.3 where the BAM images of
LuPc2:DODAB and LuPc2:DODAB/SAuNP films are compared.
103
Figure 4.3. BAM images of a) LuPc2:DODAB (1:1) and b) LuPc2:DODAB/SAuNP

(1:1/100).
4.3.2- Langmuir-Blodgett films characterization

The Langmuir films were transferred onto solid substrates. The structure
of the films deposited on a copper grid coated with carbon film was
examined using TEM (Figure 4.4). TEM images showed domains at
micrometer scale formed by LuPc2:DODAB. The gold cores of the
nanoparticles were observed as bright spots of 3.5-5.0 nm that
coincided well with the core diameter of the SAuNPs used in this work.
The SAuNPs were homogeneously distributed on the surface and were
preferentially distributed in the border between LuPc2:DODAB domains.
Figure 4.4. TEM image of the LuPc2:DODAB/SAuNP (1:1/100).
104
The UV-Vis absorption spectra of the LuPc2:DODAB films deposited onto

quartz substrates were characterized by a Q band at 665 nm and a Soret
band at 340 nm produced by transitions of the phthalocyanine ring
(Figure 4.5). These values were in good accordance with the results
reported for LuPc2 and arachidic acid mixtures28.
The intensity of the bands increased with the proportion of LuPc2 (1:10 >
1:1 > 10:1) in accordance with the Lambert-Beer´s Law. The increase in
the LuPc2 proportion also caused a shift of the Q band to higher
wavelengths (660 nm for the mixture 1:10, 665 nm for 1:1 and 675 nm
for 10:1) indicating a higher level of aggregation and an extension of the
aromaticity. The presence of J-aggregates in head-to-tail arrangement of
transition dipoles in the LB film was responsible for this effect33. When
inserting SAuNP particles in the structures, the Q and Soret bands
increased their intensity. This enhancement was proportional to the
amount of SAuNP introduced in the films, suggesting that the
amplification was caused by the interaction between the phthalocyanine
and the SAuNPs (the plasmon resonance of the SAuNP overlaps with the
Q band of the LuPc234). In addition, the magnification caused by the
SAuNPs increased proportionally with the proportion of DODAB used.
This fact could be explained taking into account that DODAB decreased
the degree of aggregation of the tightly packed phthalocyanine
molecules, facilitating the interaction between LuPc2 and the SAuNP.
Figure 4.5. UV-Vis spectra of 20 monolayers LB films of LuPc2:DODAB (1:1)

(green line), LuPc2:DODAB/SAuNP (1:1/100) (blue line) and
LuPc2:DODAB/SAuNP (1:1/500) (red line).
105
The LB films were investigated using micro-Raman technique, which

combines morphological and chemical information at micrometer scale
by coupling an optical microscope to a Raman spectrograph. Raman
spectra recorded with the 633 nm line are illustrated in Figure 4.6. Since
the LuPc2 absorbs the 633 nm laser line used here, the Resonance
Raman Scattering (RRS) should take place35.
The RRS displayed the well-known peaks corresponding to the active

vibration modes of the LuPc2 where the bands in the resonance region
were clearly enhanced. For this reason, the spectra were dominated by
four intense bands at 676 cm-1, 737 cm-1, 775 cm -1 and 813 cm-1
(associated with the Pc ring) whereas, the bands at higher wavelengths
(associated with the pyrrole stretching at 1300-1500 cm-1) were very
weak. In contrast, when using the 532 nm laser line the bands in the
1300-1600 cm-1 region associated with the pyrrole were clearly
observed36.
Figure 4.6. Resonant Raman Spectra registered using the 633 nm laser line. a)
3D line Raman mapping for a 10 monolayers LB film of LuPc2:DODAB/SAuNP
(1:1/100). b) Raman spectra collected from LuPc2:DODAB (1:1) (green line);
LuPc2:DODAB/SAuNP (1:1/100) (blue line) and LuPc2:DODAB/SAuNP (1:1/500)
(red line).
The Raman spectra given in Figure 4.6.a was obtained point-by-point for
a line of 100 µm with step of 1 µm forming a line mapping. These RRS
spectra revealed that the films were highly homogeneous with a regular
distribution of the phthalocyanine molecules. However, as observed in
Figure 4.6.b the surface-enhanced resonance Raman phenomenon
106
(SERRS)37 was not clearly observed and the presence of SAuNP only
caused an increase in the fluorescence. The reason of this observation
can be attributed to the small size and the functionalization of the
AuNPs. It has been demonstrated that the surface plasmon resonance of
small AuNPs is damped when compared to larger particles34,38. In
addition, the distance dependence of the SERRS effect with the distance
nanoparticle-molecule could also be the reason37.
4.3.3- Electrochemical characterization.

Cyclic voltammetry of indium tin oxide (ITO) electrodes modified with
Langmuir-Blodgett films of LuPc2:DODAB immersed in 0.1 mol·L-1 KCl (at
a scan rate of 0.1 V·s-1) showed a quasi-reversible process with the
anodic peak at ca. 0.7 V and the cathodic peak at ca. 0.5 V. These peaks
were due to the oxidation/reduction of the phthalocyanine ring
(LuPc2/LuPc2+) and their intensity increased with the proportion of
LuPc239. The insertion of SAuNP reduced the overpotential, causing a
shift of the oxidation peak to lower potential values while increasing the
intensity. This is illustrated in Figures 4.7.a and 4.7.b where the
voltammograms obtained from LuPc2:DODAB(10:1) and a
LuPc2:DODAB/SAuNPs (10:1/500) at different scan rates (from 0.01 to
1.0 V·s-1) are presented. For instance, at a scan rate of 1.0 V·s-1, the peak
that appeared at 0.9 V in LuPc2:DODAB moved to 0.77 V in
LuPc2:DODAB/SAuNP (10:1/500) while the intensity increased from 460
A to 920 A.
In all the LB films studied (using different proportions of LuPc2:DODAB

and containing or not SAuNPs) a linear relationship was found between
the peak current and the square root of scan rate (1/2) in the range from
0.01 to 1.0 V·s-1 indicating a diffusion limited process (Figures 4.7.c and
4.7.d). In all cases, regression coefficients were higher than 0.994 for the
oxidation peaks and higher than 0.987 for the reduction peaks.
It is worth noting that curves were steeper for the films containing
SAuNPs. For instance, the slope of the curve of the LuPc2:DODAB (1:1)
was 10.5 and the slope found in LuPc2:DODAB/SAuNP (1:1/500) was
19.3. This means that the transfer process was two times faster in the
107
presence of SAuNPs. This result proved that the mixed films containing
both LuPc2 and SAuNPs showed an excellent dynamic character in the
redox process, and also that the charge transfer within the LB film
and/or through the electrode interface was facilitated.
Figure 4.7. Cyclic voltammograms of a) LuPc2:DODAB (10:1), b)

-1
LuPc2:DODAB/SAuNPs (10:1/500) registered at scan rates from 0.01 to 1.0 V·s ,
c) Plot of Ia versus  and d) Plot of Ic versus  .
1/2 1/2
The oxidation and reduction of the Pc ring was counterbalanced by the

diffusion of ions inside the LB films. Voltammograms carried out in
electrolytes with common anions but different cations (i.e. in KCl and
MgCl2) were almost identical. In contrast, when changing the anion,
important changes were observed. Experiments carried out in the
presence of different electrolytes (Figure 4.8) demonstrated that anions
(instead of cations) diffused inside the films to maintain the
electroneutrality. For this reason, voltammograms obtained in
electrolytes containing a common anion were superimposable, whereas
108
voltammograms obtained from solutions differing in the anion were

different.
Figure 4.8. Cyclic voltammograms of LuPc2:DODAB/SAuNP (1:1/100) immersed

in KCl (black line), MgCl2 (blue line), PBS (green line) and KClO4 (red line).
4.3.4- Electrocatalytic properties. Voltammetric sensors

The electrocatalytic properties of the LuPc2:DODAB and
LuPc2:DODAB/SAuNP LB films were tested towards hydroquinone. In
order to avoid the polymerization of the hydroquinone, experiments
were carried out in the range -0.5 V to 0.7 V in buffer phosphate (PBS).
Under these conditions, the oxidation of the phthalocyanine was no
longer observed, and the redox processes registered, correspond to the
oxidation/reduction of the hydroquinone. It has to be pointed out that
the response registered when using a bare ITO electrode was extremely
weak and the peaks could not be clearly observed. As shown in Figure
4.9.a, LB films of LuPc2:DODAB 10:1 and 1:1 showed electrocatalytic
activity for oxidation of hydroquinone, illustrated by a large increase in
the intensity of the peaks associated with the oxidation/reduction of the
antioxidant and a shift of the anodic peak to lower potentials (that
appear at ca. 0.4 V). Films containing low concentrations of the
bisphthalocyanine LuPc2:DODAB (1:10) displayed a weak response
similar to that observed when using a bare ITO glass electrode. For this
reason, the LuPc2:DODAB (1:10) films were not included in further
studies.
109
Taken into account that the electrocatalytic activity was clearly

enhanced by increasing the proportion of LuPc2 in the films, such
behaviour can be attributed to the interaction of the hydroquinone with
LuPc2 40.
-1
Figure 4.9. Cyclic voltammograms registered in 40 µmol·L of hydroquinone
using: a) LuPc2:DODAB (10:1) (green line), LuPc2:DODAB (1:1) (red line) and
LuPc2:DODAB (1:10) (blue line). b) LuPc2:DODAB (1:1) (green line),
LuPc2:DODAB/SAuNPs (1:1/100) (blue line) and LuPc2:DODAB/SAuNPs (1:1/500)
-1
(red line). Scan rate, 0.1 V·s .
The presence of SAuNPs was found to increase the electrocatalytic

efficiency of the sensors towards oxidation of hydroquinone. This
increase was proportional to the concentration of SAuNPs (Figure 4.9.a).
For instance, the electrocatalytic efficiency of the films was increased by
5% for LuPc2:DODAB/SAuNPs (1:1/100) and 10% for
LuPc2:DODAB/SAuNPs (1:1/500) (based on the current intensity of
LuPc2:DODAB measured at 0.4 V).
The presence of SAuNPs increased the intensity of the oxidation current

by creating active sites and improving the communication within the
phthalocyanine films and the electrode. Similar electrocatalytic effects
have been observed in electrodes covered with metallic nanoparticles
and other organic materials41,42.
As shown in the previous paragraphs, the peak-to-peak separation and

electron transfer kinetics of LuPc2:DODAB/SAuNP modified electrodes
110
changed with the amount of SAuNP inserted in the LB structure.

However, it is important to notice that an excess of SAuNP (injecting
more than 500 l of SAuNP under the water subphase) did not produce
further improvement leading to a decrease in electrochemical sensitivity
and irreproducibility.
The peak current (measured at the voltage of oxidation or reduction of

hydroquinone) showed a root square dependence with the scan rate
(from 0.01 to 1.0 V·s-1) which indicated diffusion controlled mechanism
for the oxidation/reduction of the species in solution. As shown in Table
4.2 that collects the data corresponding to the anodic waves, the
presence of SAuNPs increased the slope of the curve pointing to a
favoured charge transfer process. The increase varied between 15% and
22% depending on the sensor.
2
Table 4.2. Slope and regression coefficient (R ) of the curves obtained by
representing the intensity of the anodic peak of the oxidation of hydroquinone
-1
(at 0.45 V) versus the square root of the scan rate (from 0.01 to 1.0 V·s ).
E=0.45V
Sensor
Slope R2
LuPc2:DODAB (1:1) 7.3847 0.999
LuPc2:DODAB/SAuNP (1:1/100) 7.9500 0.999
LuPc2:DODAB/SAuNP (1:1/500) 8.9535 0.999
LuPc2:DODAB (10:1) 7.4009 0.998
LuPc2:DODAB/SAuNP (10:1/100) 7.8023 0.998
LuPc2:DODAB/SAuNP (10:1/500) 8.4641 0.992
The intensity of the peaks associated with the phenol increased linearly
with the concentration of hydroquinone in the range from 5 µmol·L-1 to
150 µmol·L-1 (Figure 4.10). The calibration curves were constructed by
representing the intensity value of the anodic (at ca. 0.45 V) and
cathodic (at ca. -0.1 V) waves versus the hydroquinone concentration.
The limits of detection (LOD) were obtained according to the 3sd/m
criterion, where m is the slope of the calibration curve, and sd was
estimated as the standard deviation (n=7) of the voltammetric signal
111
registered using the blank. The LOD calculated for all the sensors
prepared are collected in Table 4.3. The improvement of the sensitivity
caused by the synergistic electrocatalytic effect allowed the
LuPc2:DODAB/SAuNP sensors to reach limits of detection in the range of
10-6-10-7 mol·L-1 which are lower than those obtained for the
corresponding LuPc2:DODAB electrodes. These results demonstrate that
the LuPc2:DODAB/SAuNP nanostructured sensors can be used to
quantify the presence of hydroquinone in the range usually present in
foods and even below.
The responses were highly reproducible with a coefficient of variation

(n=10) lower than 2.0% between similar sensors (same composition).
The reproducibility intra-assay was lower than 1.0%.
Figure 4.10. a) Response LuPc2:DODAB/SAuNPs 1:1/100 towards hydroquinone

(4-150 µmol·L in PBS). b) Plot of Ia () or Ip () versus  .
-1 1/2
4.4 CONCLUSIONS
In this chapter, a method to prepare nanostructured electrodes of two
electrocatalytic materials (functionalized AuNPs and lutetium
bisphthalocyanine) using the Langmuir-Blodgett technique has been
developed.
The surface pressure-mean molecular area isotherms combined with

BAM images indicated that SAuNPs were adsorbed into the floating films
formed by LuPc2 and DODAB. Spectroscopic and microscopic studies
confirmed that homogeneous LB films can be transferred to solid
substrates.
112
Electrocatalytic properties of the LuPc2:DODAB/SAuNP electrodes

towards the detection of hydroquinone were evidenced by a decrease in
the oxidation potential at which oxidation of hydroquinone takes place.
It was demonstrated that the SAuNPs and LuPc2 have a synergistic effect
in terms of improving electrocatalysis for the detection of hydroquinone
as confirmed by the enhancement of the oxidation peak current (at 0.45
V) and the reduction peak current (at -0.01 V).
Voltammetric responses were proportional to the concentration of

hydroquinone between 5 µmol·L-1 to 150 µmol·L-1, and the limit of
detection was in the range of 10-6-10-7 mol·L-1 with good reproducibility.
113
Reduction Oxidation
Sensor
LOD (mol·L-1) R2 Slope E (V) LOD (mol·L-1) R2 Slope E (V)

LuPc2:DODAB (1:1) 2.14x10-6 0.989 -0.4231 -0.03 1.65x10-6 0.989 0.4471 0.52
LuPc2:DODAB/SAuNP
5.60x10-7 0.990 -0.4732 -0.07 9.30x10-7 0.989 0.5055 0.52
(1:1/100)
LuPc2:DODAB/SAuNP
114
4.60x10-7 0.983 -0.4872 -0.08 3.30x10-7 0.989 0.5254 0.52
(1:1/500)
LuPc2:DODAB (10:1) 2.71x10-6 0.994 -0.4772 -0.08 1.74x10-6 0.992 0.5699 0.42
LuPc2:DODAB/SAuNP
7.80x10-7 0.995 -0.3207 -0.08 1.41x10-6 0.998 0.515 0.48
(10:1/100)
LuPc2:DODAB/SAuNP
LuPc2:DODAB and LuPc2:DODAB/SAuNP nanostructured films.
8.77x10-7 0.9892 -0.2399 -0.01 1.05x10-6 0.991 0.368 0.60

(10:1/500)
Table 4.3. Detection limits calculated for the detection of hydroquinone using
References
1. Bartosz, G., Food oxidants and antioxidants: chemical, biological, and
functional properties, 2013. Ed. CRC press, Florida (USA).
2. Makhotina, O., Kilmartin, P.A. Anal. Chim. Acta. 2010, 668, 155-165.
3. Prehn, R., Gonzalo-Ruiz, J., Cortina-Puig, M. Curr. Anal. Chem. 2012, 8,

472-484.
4. Narducci, D. Sci. Adv. Mat. 2011, 3, 426-435.
5. Chen, A.C., Chatterjee, S. Chem. Soc. Rev. 2013, 42, 5425-5438.
259.
7. Zhou, Y.G., Campbell, F.W., Belding, S.R., Compton, R.G. Chem. Phys.
Lett. 2010, 497, 200-204.
8. Pena-Pereira, F., Duarte, R.M.BO., Duarte, A.C. Trends Anal. Chem.

2012, 40, 90-105.
9. Shan, X., Díez-Pérez, I., Wang, L., Wiktor, P., Gu, Y., Zhang, L., Wang,
W., Lu, J., Wang, S., Gong, Q., Li, J., Tao, N. Nature Nanotech. 2012, 7,
668-672.
10. Jiang, J., Functional Phthalocyanine Molecular Materials. Series:

Structure and Bonding 13, 2010. Ed. Springer, Berlin (Germany).
11. Rodriguez-Mendez, M.L. Sensing properties of phthalocyanines, in:

Encyclopedia of Sensors, 2006, 111-134. Ed. America Scientific
Publishers, California (USA).
12. Zagal, J.H., Griveau, S., Silva, J.F., Nyokong, T., Bedoui, F. Coord.
Chem. Rev. 2010, 254, 2755-2794.
13. Bouvet, M., Gaudillat, P., Suisse, J.M. J. Porphyrins. Phthalocyanines.

2013, 17, 913-919.
115
14. Sorokin, A.B., Kudrik, E.V. Catal. Today. 2011, 159, 37-46.
15. Valli, L. Adv. Colloid. Interface. Sci. 2005, 116, 13-44.
16. Rodriguez-Mendez, M.L., Gay, M., de Saja, J.A. J. Porphyrins.

17. Volpati, D., Alessio, P., Zanfolim, A.A., Storti, F.C., Job, A.E., Ferreira,
M., Riul, A., Oliveira, O.N.Jr., Constantino, C.J.L. J. Phys. Chem. B. 2008,
112, 15275-15282.
18. Pavinatto, F.J., Fernandez, E.G.R., Alessio, P., Constantino, C.J., de

Saja, J.A., Zucollotto, V., Oliveira, O.N. Jr., Apetrei, C., Rodriguez-Mendez,
M.L. J. Mat. Chem. 2010, 21, 4995-5003.
19. Nyokong, T., Antunes, E. Coord. Chem. Rev. 2013, 257, 2401-2418.
20. Chauke, V.P., Chidawanyika, W., Nyokong, T. Electroanal. 2011, 23,

487-496.
21. Hone, C., Walker, P.I., Evans-Gowing, R., Fitzgerald, S., Beeby, A.,
Chambrier, I., Cook, M.J., Russell, D.A. Langmuir. 2002, 18, 2985-2987.
22. Pal, M., Ganesan, V. Langmuir. 2009, 25, 13264-13272.
23. Alencar, W.S., Crespilho, F.N., Martins, M.V.A., Zucolotto, V.,

Oliveira, O.N.Jr., Silva, W.C. Phys. Chem. Chem. Phys. 2009, 11, 5086-
5091.
24. Aoki, P.H.B., Alessio, P., de Saja, J.A., Constantino, C.J.L. J. Raman
Spectrosc. 2010, 41, 40-48.
25. Comeau, K.D., Meli, M.V. Langmuir. 2012, 28, 377-381.
26. Deisinger, P.J., Hill, T.S., English, J.C. J. Toxicol. Environ. Health. 1996,
47, 31-46.
27. Linaje, M., Quintanilla, M.C., Gonzalez. A., Valle, J.L., Alcaide, G.,
Rodriguez-Mendez, M.L. Analyst. 2000, 125, 341-346.
116
28. Clavijo, R.E., Battisti, D., Aroca, R., Kovacs, G.J., Jennings, C.A.
Langmuir. 1992, 8, 113-117.
29. Maitrot, M., Guillaud, G., Boudjema, B., Andre, J.J., Strzelecka, H.,
Simon, J., Even, R. Chem. Phys. Letters. 1987, 133, 59-62.
30. Souto, J., Tomilova, L., Aroca, R., de Saja, J.A. Langmuir. 1992, 8, 942-
946.
31. Viseu, M.I., Gonçalves da Silva, A.M., Antunes, P., Costa, S.M.B.
Langmuir. 2002, 18, 5772-5781.
32. Martin, M.T., Möbius, D. Thin Solid Films. 1996, 284-285, 663-666.
33. Rodriguez-Mendez, M.L., Gorbunova, Y., de Saja, J.A. Langmuir.

2002, 18, 9560-9565.
34. Huang, X., El-Sayed, M.A., J. Advanced. Research. 2010, 1, 13–28.
35. Moskovits, M. J. Raman Spectrosc. 2005, 36, 485-496.
36. Rodriguez-Mendez, M.L., Aroca, R. Spectrochim. Acta A. 1993, 49,

965-973.
37. Aroca, R., Pieczonka, N., Kam, A.P. J. Porphyrins Phthalocyanines.

2001, 5, 25-32.
38. Le Ru, E.C., Etchegoin, P.G., Meyer, M. J. Chem. Phys. 2006, 125,
204701.
39. Fernandes, E.G.R., Brazaca, L.C., Rodriguez-Mendez, M.L., de Saja,

40. Gay Martin, M., Rodriguez-Mendez, M.L., de Saja, J.A. Langmuir.

2010, 26, 19217-19224.
41. Azad, U.P., Ganesan, V. Electroanal. 2010, 22, 575-583.
42. Pal, M., Ganesan, V. Langmuir. 2009, 25, 13264-13272.
117
CHAPTER 5
Bioelectronic Tongue Based on Lipidic
Nanostructured Layers Containing Phenol
Oxidases and Lutetium Bisphthalocyanine for the
Analysis of Grapes
When complex samples (food and beverages) are analyze using

voltammetric sensors, the responses obtained are not directly related to
the concentration of a specific analyte or family of molecules. The
information under the curves is associated with every redox active
species present on the sample. Constructing arrays of sensors is a way to
improve the electrochemical analysis. The cross-selectivity attained
when more than one sensor is used on the analysis allows discriminating
samples with similar characteristics (e.g. varieties of grapes, types of
wines…). For this purpose an exhaustive data treatment is needed
including pre-processing of the data, training of the system, prediction,
validation…
In order to make sensors more specific, enzymes can be included in the

sensitive layer. As stated in the introduction enzymes are proteins that
catalyze specifically some reactions which involve molecules that can be
bind to their active site. Phenol oxidases (tyrosinase and laccase) are a
family of enzymes that catalyze reactions associated with polyphenols
redox processes.
Langmuir-Blodgett (LB) technique is of special interest in the field of

biosensors because it allows immobilizing the enzyme in a biomimetic
environment, helping to preserve the activity of the enzyme, and co-
immobilize the electron mediator.
In this work, biosensors containing phenol oxidases have been

developed using LB technique. The proteins have been incorporated into
LB films of arachidic acid (AA), simulating biological membranes, using
lutetium bisphthalocyanine (LuPC2) as electron mediator. Cyclic
voltammetry (CV) has been applied to detect six phenolic compounds
including one monophenol, three diphenols and two triphenols. The
complementarity achieved by the different sensors and the electron
mediator capability of the bisphthalocyanine will be discussed. In
addition, this bioelectronic tongue will be employed to analyze must
from different varieties of grapes and its capability of discrimination will
be analyzed. These results have been published in Biosensors and
Bioelectronics.
121
5.1 INTRODUCTION
The determination of phenols, the main antioxidants in foods, has been
widely investigated using traditional techniques including spectroscopy,
chromatography and electrochemical methods1,2.
A promising approach in food analysis consists in the use of electronic

tongues (ET) which are multisensor systems based on a number of low-
selective sensors and use advanced mathematical procedures for
processing the electrochemical signals, based on pattern recognition
and/or multivariate data analysis3,4. Electronic tongues provide global
information about the sample instead of information about specific
compounds.
Electrochemical sensors are the most widely used sensing units in

electronic tongues. They include potentiometric5, amperometric6,
voltammetric7-9 or impedimetric sensors10.
Arrays of voltammetric electrodes chemically modified with

electroactive materials (e.g. phthalocyanines) have demonstrated to be
particularly interesting for the analysis of phenolic compounds11-13.
When using such electrodes, voltammograms show redox peaks
produced by the electrode material and by the solution. In addition, the
interactions between the electrode and the solution (i.e. electrocatalytic
activity of the sensing material) produce shifts in the peak positions and
changes in their intensity. In this way, each electrode produces a distinct
response towards different solutions. The intrinsic complexity, richness
and cross-selectivity of the signals generated by an array of
voltammetric electrodes are an advantage because each curve contains
large amount of information about the sample7,9 . Phthalocyanines (MPc)
and their sandwich type lanthanide derivatives (LnPc2) are among the
most suitable materials for electrochemical sensors due to their well-
known electrocatalytic properties9,13,14. They have demonstrated to
behave as excellent modifiers for the detection of a variety of analytes
including polyphenolic compounds11,12,15. Nanostructured electrochemical
sensors based on phthalocyanines can be prepared using the Langmuir-
Blodgett (LB) technique16,17.
122
On the other hand, electrochemical biosensors are an interesting

alternative for the analysis of phenols due their high sensitivity and
selectivity. They contain phenol oxidase enzymes such as tyrosinase or
laccase combined with appropriate electron mediators such as metallic
nanoparticles, graphene and conducting polymers among others18. It has
been demonstrated that MPcs and LnPc2 can also be used as electron
mediators in tyrosinase biosensors19-21. The LB technique is of special
interest in the field of biosensors because using this method, enzymes
can be immobilized in a nanostructured lipidic layer with a structure
similar to that of the biological membranes. This biomimetic
environment can help to preserve the functionality of the enzyme20-22. In
addition, using the LB technique, the enzyme and the electron mediator
can be co-inmobilized in a single sensitive layer, facilitating the electron
transfer between the enzyme and the electrode.
Some attempts have been carried out to develop arrays of biosensors

containing phenol oxidases for the detection of phenols (the so-called
bioelectronic tongues)23,24. It has been demonstrated that arrays of
biosensors combine the advantages of classical arrays of electrochemical
sensors that provide global information about the sample, with the
specificity of the enzyme-substrate reaction typical of biosensors.
5.2.1- Chemicals
All chemical and solvents were of reagent grade. Deionized water
(resistivity of 18.2 M·cm-1) was used to prepare subphases and
solutions.
Laccase, from Trametes versicolor (EC Number: 1.10.3.2, activity of 20.7

U·mg-1) and Tyrosinase (from mushroom EC 232-653-4), noted activity of
3610 U·mg-1 were purchased from Sigma Chemical. 70 g·mL-1 solutions
of tyrosinase and laccase were prepared in buffer phosphate 0.01
mol·L-1 (pH=7.0) (PBS).
123
The lutetium (III) bisphthalocyaninate (LuPc2) was synthesized following

a previously published procedure25.
5.2.2- Langmuir and Langmuir-Blodgett films characterization

Isotherms and LB films were prepared in a KSV 2000 Langmuir-Blodgett
trough (KSV Instruments, Finland) equipped with a Wilhelmy plate to
measure the surface pressure.
According to a previously published method, LB films were prepared

using a PBS-NaCl subphase (NaCl 0.1 mol·L-1, phosphate buffer 0.01
mol·L-1 of pH=7.0 in ultrapure water)21.
Mixed films containing arachidic acid (AA) and lutetium

bisphthalocyanine (LuPc2) were prepared by spreading 250 µl of a
mixture 10:1 (AA/LuPc2) dissolved in chloroform (110-5 mol·L-1) onto
the PBS-NaCl subphase. The surface-area isotherms were measured by
compressing the floating molecules at a speed of 10 mm·min-1.
At a surface pressure of 40 mN·m-1, 20 monolayers were deposited onto

previously cleaned ITO glass surface by Y-type deposition with a transfer
ratio close to 1.
LB films containing enzyme, arachidic acid and lutetium

bisphthalocyanine (Enz/AA/LuPc2) were prepared in two steps. First, 10
monolayers of AA/LuPc2 were deposited using the method described in
the previous paragraphs. Then, 10 monolayers of Enz/AA/LuPc2 were
deposited onto the AA/LuPc2 layers as follows: 250 µl of the AA/LuPc2
solution were spread onto the PBS-NaCl subphase. When the solvent
was evaporated, 100 µl of a 70 µg·ml-1 solution of the corresponding
enzyme in 0.01 mol·L-1 PBS were injected drop by drop underneath the
air/liquid interface. Barriers were compressed at a speed of 10 mm·
min-1. At a surface pressure of 40 mN·m-1, 10 monolayers of
Enz/AA/LuPc2 were deposited onto ITO glass with a substrate speed of 3
mm·min-1. Films were built by Y type deposition with a transfer ratio
close to 1.
124
After preparation, LB films of Enz/AA/LuPc2 were treated with

glutharaldehyde to form covalent bonds between the enzymes and the
amphiphilic molecules26.
Langmuir films were analyzed with Brewster Angle Microscopy (BAM)

using a KSV MicroBAM.
AFM images were registered in LB films deposited onto ITO using a

MultiMode Scanning Probe Microscope Model MMAFM-2 from Digital
Instruments.
5.2.3- Electrochemical measurements

The electrochemical measurements were carried out in an EG&G
PARSTAT 2273 potentiostat/galvanostat using a conventional three-
electrode cell. The LB films were used as working electrode. The
reference electrode was Ag|AgCl/KCl 3 M and the counter electrode was
a platinum plate. Cyclic voltammograms were registered at a sweep rate
of 0.1 V·s-1.
5.2.4- Phenols and grapes

10-3 mol·L-1 stock solutions of phenolic compounds including one
monophenol (vanillic acid), two orto-diphenols (catechol and caffeic
acid), one para-diphenol (hydroquinone) and two triphenols (gallic acid
and pyrogallol) were prepared by solving the corresponding compound
in PBS. Solutions with lower concentration were prepared by dilution.
Grapes of five different varieties (Tempranillo, Garnacha, Cabernet-

Sauvignon, Prieto Picudo and Mencía) were harvested in 2012 in the
Castilla y León region (Spain) by the Agrotechnological Institute of the
regional Government (ITACYL), and by a cellar of the region (Bodega
Cooperativa de Cigales). The Enological Centre of Castilla y León carried
out the chemical analysis including the Total Polyphenol Index (TPI)
following international regulations27.
5.2.5- Statistical analysis. Data treatment

A non-supervised multivariate method, Principal Component Analysis
(PCA), was used to analyze the voltammetric curves and to evaluate the
125
capability of discrimination of the array of sensors. The voltammetric

curves were mathematically pre-processed. After normalization a
windowed slicing method was used to reduce the number of data per
sample28. Using this method, 10 parameters per curve were obtained and
used as a data source for statistical analysis. All computations and
chemometric analysis were carried out using the software Matlab v5.3.
5.3.1- Langmuir monolayers and Langmuir-Blodgett films

The isotherm of AA/LuPc2 registered in PBS-NaCl subphase is shown in
Figure 5.1.a. The limiting area calculated from the slope of the curve was
37 Å2. This value corresponds to an edge-on orientation with the Pc ring
tilted and assembled with the main molecular axis parallel to the water
subphase (lying flat they should occupy an area approximately 20 Å20
Å)29.
Tyrosinase and laccase could be effectively adsorbed into the non-

compressed AA/LuPc2 films as demonstrated by the increase in the
surface pressure that occurred upon injection of the enzymes
underneath the air/liquid interface. In the case of tyrosinase, a constant
and progressive increase in the surface pressure was observed. After 45

minutes (when the surface pressure reached a value of ca. 0.35 mN·m-1)
a plateau was attained. The changes in pressure due to the
incorporation of laccase to the non-compressed film followed a different
kinetics. The increase in the surface pressure started 30 minutes after
the injection and reached a plateau 30 minutes later at 0.40 mN·m-1
(total time for saturation was 60 minutes). Once the surface pressure
was stabilized, the floating films were compressed at a constant speed
of 10 mm·min-1. Under these conditions reproducible isotherms were
obtained (Figure 5.1.a).
The shapes of these isotherms differed from that of AA/LuPc2. The

formation of the bidimensional solids started at higher areas per
molecule and the limiting areas per molecule were larger than the
126
values observed for the AA/LuPc2 Langmuir film, confirming the

adsorption of the enzymes into the films. The different areas per
molecule observed for Tyr/AA/LuPc2 (51 mN·m-1) and Lac/AA/LuPc2 (47
mN·m-1), can be attributed to the different size and molecular weight of
both enzymes (120 kDa for tyrosinase and 50-100 kDa for laccase).
The adsorption of the enzyme into the floating films was also confirmed
by the dependence of the limiting area with the time elapsed between
the injection of the enzyme and the beginning of the compression. This
is also illustrated in Figure 5.1.a, where isotherms registered by waiting
30 and 60 minutes between the injection of the enzyme and the
beginning of the compression are compared. In the case of
Tyr/AA/LuPc2, the limiting area increased with the time elapsed after
injection (from 0 to 30 minutes), but isotherms registered after waiting
30 or 60 minutes were almost identical. The variation was dramatic for
the Lac/AA/LuPc2. This result is in good accordance with the different
adsorption kinetics mentioned in the previous paragraphs.
Compression-expansion cycles were also studied. Upon decompression,

AA/LuPc2 floating monolayers showed a certain hysteresis, but
successive compression-expansion cycles where highly reproducible
(Figure 5.1.b). Hysteresis was also observed in monolayers containing
tyrosinase (Figure 5.1.c) or laccase (Figure 5.1.d). The isotherms of the
decompression were similar to those of AA/LuPc2 film, pointing to
desorption of the enzyme during the expansion cycle. During the second
cycle, Tyr/AA/LuPc2 film recovered the original shape, but the second
compression of Lac/AA/LuPc2 was similar to the isotherm of the
AA/LuPc2 film. This means that after desorption, tyrosinase could be
partially reinserted, whereas laccase could not be readsorbed into the
film. It can be concluded that the insertion of laccase into the floating
film is more difficult than the insertion of tyrosinase. This could seem
surprising because the size of tyrosinase is considerably larger than that
of laccase. However, the hydrophobic environment provided by the
floating film is more favorable for the tyrosinase which is a
127
transmembrane protein and only a small part of the enzyme extends in

the cytoplasmic hydrophilic environment.
Figure 5.1. a) Pressure Area isotherms registered in a PBS-NaCl subphase after

different waiting times. Compression-expansion cycles of b) AA/LuPc2, c)
Tyr/AA/LuPc2 and d) Lac/AA/LuPc2.
The quality of the floating films was further analyzed with Brewster
Angle Microscopy (BAM). The homogeneity of the Langmuir films
containing tyrosinase or laccase was similar to that of the AA/LuPc2 films
(Figure 5.2).
In order to visualize the topography of the LB films, AFM was employed.

As presented in Figure 5.2.a, AA/LuPc2 films showed a homogeneous and
smooth surface. Regarding Enz/AA/LuPc2 films, immobilized enzymes
were clearly observed showing their characteristic cloud-like
morphology (Figure 5.2.b). The enzymes were well and evenly
distributed on the surface of the electrode, and no large aggregates
were formed.
128
Figure 5.2. BAM images, AFM images and roughness profile of the sensors a)
AA/LuPc2 and b) Lac/AA/LuPc2 films.
5.3.2- Electrochemical response towards phenols

The voltammetric responses of the array formed by three electrodes
(AA/LuPc2, Tyr/AA/LuPc2 and Lac/AA/LuPc2) towards six phenolic
compounds (40 µmol·L-1) are shown in Figure 5.3. The responses were
highly reproducible with a coefficient of variation (n=10) lower than
1.5% (calculated from the peaks associated with antioxidants) and lower
than 0.75% (calculated from the peaks associated with LuPc2). The
reproducibility between different sensors, containing the same enzyme,
immersed in the same solution was always lower than 1.75%.
The electrochemical response of AA/LuPc2 nanostructured films towards

phenols (represented in Figure 5.3 as green lines) showed a quasi-
reversible and intense redox pair associated with the LuPc2-/LuPc20
process at ca. E1/2= -0.15 V. However, due to the interaction with
phenols, this peak appeared at different positions and showed diverse
intensities depending on the antioxidant analyzed. Voltammograms also
showed redox processes arisen from the oxidation/reduction of phenols.
The position and intensity of such peaks depended on the type of phenol
and were in good accordance with previously published results21,30.
129
Taking into acount that oxidation of phenols barely takes place at bare
ITO electrodes and considering that when using a carbon paste
electrode the oxidation occurs at higher potentials than those reported
here30, it can be confirmed that the AA/LuPc2 sensor reduced the
overpotential and improved the reversibility, validating the
electrocatalytic behavior of the LuPc2. For instance, the oxidation of
catechol in carbon paste electrode ocurred at 0.5 V and the reduction in
the reverse scan appeared at 0.05 V (E1/2= 0.45 V). When using
AA/LuPc2 sensor, the oxidation of the phenol appeared at 0.4 V and the
cathodic wave at 0.15 V (E= 0.25V). The intensity of the peaks was also
clearly increased.
One of the objectives of this chapter was to improve the selectivity of

the sensors by introducing enzymes in biomimetic layers containing
LuPc2 as electron mediator19,26,27. Two phenol oxidases (tyrosinase and
laccase) were selected. Tyrosinase oxidizes monophenols and o-
diphenols to the corresponding quinone, whereas laccase catalyzes the
oxidation of a larger variety of aromatic compounds such as substituted
mono- and polyphenols31.
Using this strategy, Tyr/AA/LuPc2 and Lac/AA/LuPc2 sensors were

prepared and their response towards phenols was analyzed in terms of
sensitivity and cross-selectivity. The results are shown in Figure 5.3
(Tyr/AA/LuPc2 responses are represented as blue lines and Lac/AA/LuPc2
as red lines). When biosensors containing tyrosinase or laccase are
immersed in a phenolic solution, the phenol is enzymatically oxidized to
the corresponding quinone (no need of external voltage). When the
voltage was biased to positive values, an anodic peak at ca. 0.5 V was
observed. This peak corresponded to the electrochemical oxidation of
the phenol to the quinoid form. During the reverse scan, both the
enzymatically and the electrochemically formed o-quinone molecules
were reduced simultaneously at ca. -0.2 V. This concurrent reduction
explains why this peak was more intense in Enz/AA/LuPc2 than in
AA/LuPc2 sensors. Moreover, the enzymatic processes were favored by
130
the presence of the LuPc2, which acted as electron mediator, producing

signal amplification.
Unexpectedly, the presence of the enzyme also increased the intensity

of the peaks associated with the oxidation/reduction of the
phthalocyanine ring (at ca. -0.2 V). These changes confirmed the
interaction between the phthalocyanine and the enzyme and the
subsequent electrocatalytic effect.
The most remarkable conclusion that can be extracted from Figure 5.3 is
the important degree of cross-selectivity attained by the array of
sensors. Significant differences were observed in the responses of
Tyr/AA/LuPc2 and Lac/Tyr/LuPc2 that come from their different
specificity.
For instance, as both enzymes react with monophenols, their response

towards vanillic acid was similar. Catechol is an o-diphenol and
consequently the response shown by tyrosinase was clearly more
intense than the response observed for laccase. Moreover, the
electrochemical oxidation of catechol occurred at lower potentials in
Tyr/AA/LuPc2 (0.38 V) than in Lac/AA/LuPc2 (0.62 V). Caffeic acid is also
an o-diphenol, but the presence of substituents in the benzene ring
changes the polarity and the size of the substrate, modifying the
interaction with the active site of the tyrosinase. A similar argument
applies for hydroquinone, a p-diphenol that shows higher affinity
towards laccase.
131
Figure 5.3. Cyclic voltammograms of AA/LuPc2 (green line), Tyr/AA/LuPc2 (blue

-1
line) and Lac/AA/LuPc2 (red line) immersed in 40 μmol·L PBS solutions of a)
vanillic acid, b) catechol, c) caffeic acid, d) hydroquinone, e) gallic acid and f)
pyrogallol.
The triphenols tested, gallic acid and pyrogallol, showed completely

different responses towards Tyr/AA/LuPc2 and Lac/AA/LuPc2. Gallic acid
is a substituted triphenol and according to the results obtained, its
affinity towards tyrosinase and laccase is not good because enzymatic
activity was not observed. In the case of pyrogallol, the first cycle was
clearly different from the signals obtained in the subsequent ones which
132
were reproducible and similar to those obtained when the electrodes

were immersed in catechol. The reason is that the oxidation of
pyrogallol occurs in two steps: during the first cycle, the quinoid form is
obtained and during the second cycle, the redox properties of the o-
quinone are observed. This behavior also explains that sensors modified
with tyrosinase or laccase provided different responses, due to their
different affinity towards o-diphenols.
The detection limits (LOD) were determined by measuring the responses

of the sensors towards phenol solutions with concentrations ranging
from 4 to 150 µmol·L-1. The peaks associated with phenols were clearly
identified by the progressive increase in their intensity. The detection
limit were statistically calculated using LOD=3sd/m, where sd is the
standard deviation of the blank at the potential measured and m is the
slope of the calibration curve. Peak positions, LOD and regression
coefficients are shown in Table 5.1. The LOD found in Enz/AA/LuPc2
were at least one order of magnitude lower than those found in
AA/LuPc2. They were in the range of those published for tyrosinase
biosensors containing other electron mediators32.
The dynamic behavior was examined by analyzing the effect of the

sweep rate (from 0.025 to 0.2 V·s-1) on the intensity of the voltammetric
responses. The experiments were carried out in 100 µmol·L-1 phenolic
solutions. A clear dependence of the intensity of the peaks with the
square root of the sweep rate could be noticed for all the phenols and
sensors analyzed (Figure 5.4). It is worthily noting that in the case of the
Enz/AA/LuPc2 films, the linear relationship between the peak current
and the square root of the scan rate was approximately 5 times faster
than the value found in the AA/LuPc2 electrode. This result indicates that
the charge transfer within the LB film and/or through the electrode
interface was facilitated and this improvement should be related to the
enzymatic activity.
Table 5.1. Detection limits calculated from the peaks associated with phenols
redox processes.
133
Sensor Ea (V) LD (mol·L-1) R2 Ec (V) LD (mol·L-1) R2

Vanillic Acid
AA/LuPc2 0.200 1.3310-4 0.921 0.100 6.5910-5 0.921
Lac/AA/LuPc2 0.200 5.1010-6 0.975 0.110 4.0610-6 0.976
Tyr/AA/LuPc2 0.200 7.6910-6 0.962 0.100 5.4510-6 0.975
Catechol
AA/LuPc2 0.528 4.2810-6 0.992 0.083 3.0410-6 0.994
Lac/AA/LuPc2 0.550 4.5910-7 0.990 0.065 4.8810-7 0.992
Tyr/AA/LuPc2 0.600 4.8110-7 0.992 0.060 5.1810-7 0.997
Caffeic Acid
AA/LuPc2 0.450 4.1910-6 0.997 0.100 3.4510-6 0.997
Lac/AA/LuPc2 0.520 5.8910-7 0.990 0.054 7.7410-7 0.981
Tyr/AA/LuPc2 0.500 5.7310-7 0.997 0.060 6.2310-7 0.996
Hydroquinone
AA/LuPc2 0.600 3.3410-6 0.988 0.058 3.3810-6 0.989
Lac/AA/LuPc2 0.550 5.1810-7 0.976 -0.090 5.4210-6 0.972
Tyr/AA/LuPc2 0.600 5.9410-7 0.995 -0.100 6.4010-7 0.995
Gallic Acid
AA/LuPc2 0.698 3.6910-6 0.992
Lac/AA/LuPc2 0.699 4.1010-8 0.991
Tyr/AA/LuPc2 0.700 4.9710-7 0.985
Pyrogallol
AA/LuPc2 0.250 2.6610-5 0.991 0.528 2.1410-6 0.989
Lac/AA/LuPc2 0.360 1.8710-6 0.998 0.699 3.0510-7 0.998
Tyr/AA/LuPc2 0.200 3.4110-6 0.992 0.500 4.5110-7 0.992
134
Figure 5.4. Correlation between the intensity of a) anodic and b) cathodic peaks
versus the square root of the scan rate ( ): () AA/LuPc2, () Tyr/AA/LuPc2
1/2
-1
and () Lac/AA/LuPc2, immersed in caffeic acid 100 µmol·L .
5.3.3- Array of sensors. Discrimination of phenolic compounds

The intrinsic complexity and cross-selectivity of the signals generated by
the array of voltammetric electrodes could be used to discriminate the
phenols using Principal Component Analysis (PCA). The high amount of
information displayed by the voltammograms, makes difficult the data
analysis, increasing the processing time. A pre-treatment method of the
voltammetric curves has been developed in our laboratory that allows
reducing the number of variables to a few representative values13,28. As a
result of the calculations, 10 input variables were extracted for each
voltammetric curve and used for PCA calculations. As shown in PCA
scores plot (Figure 5.5.a) PC1, PC2 and PC3 explained 98% of the total
variance between the samples. The separated clusters indicated that the
six phenols could be clearly discriminated. The graph could be divided
into three regions corresponding to the chemical structure of the
studied phenols: the monophenol (vanillic acid) appeared in the region
of positive PC1. The three diphenols (catechol, caffeic acid and
hydroquinone) appeared in the left part of the graph in the negative PC1
region. Finally, gallic acid and pyrogalloll (triphenols) were located in the
middle region of PC1 and PC2 but in positive PC3 values.
Figure 5.5.b shows the contribution of the variables (10 kernels per
sensor) in a two-dimensional PCA loading plot. As observed in the figure,
all the variables used in the PCA analysis showed high loading
parameters for a particular principal component, indicating a positive
influence in the discrimination process. Moreover, Figure 5.5.b shows
that the variables associated with each sensor appear in different
regions of the plot.
135
Figure 5.5. a) PCA scores plot of the considered phenols. Yellow- Vanillic acid,
Purple- Hydroquinone, Pink- Caffeic acid, Green- Catechol, Orange- Pyrogallol
and Blue- Gallic acid. b) Corresponding loadings plot in terms of PC1 versus PC2.
5.3.4- Array of sensors. Response towards grapes

The array of sensors and biosensors described in the previous sections
was exposed to musts prepared from grapes of different varieties
(diluted 50% in water). Voltammograms were dominated by the redox
response of the phenolic groups present in musts that appeared as
anodic peaks in the 0.4-0.8 V region, and the corresponding cathodic
waves in the 0.35 V region. The intensities and positions of those peaks
were related to the total polyphenol index (TPI) measured by chemical
methods that, in turn, depended on the grape variety. The responses
were highly reproducible with a coefficient of variation (n=7) always
lower than 4.5%. Obviously in such a complex media, the peaks were
broader than in the phenolic solutions and a variety of other small and
not well-defined peaks were observed. In summary, each electrode
provided a different response towards the same must sample and an
important degree of cross-selectivity was attained.
136
Figure 5.6. a) PCA scores plot of the different varieties of grapes: T-Tempranillo
(orange), G-Garnacha (blue), C-Cabernet-Sauvignon (pink), M-Mencía (purple)
and PP-Prieto Picudo (green). b) Corresponding loadings plot in terms of PC1
versus PC2.
As observed in the PCA scores plot, the array was able to discriminate
grapes according to the grape variety (Figure 5.6.a). PC1, PC2 and PC3
explained 92% of the total variance between musts. The position of the
clusters was related to the chemical composition of the grape juices. The
third Principal Component PC3, was responsible of the discrimination of
the samples according to their Total Polyphenol Index: The must
obtained from the variety Mencia (TPI of 12) appeared in the top part of
the graph in the PC3 positive values, whereas the must obtained from
Tempranillo (TPI of 20) was located in the lower part of the figure. Musts
obtained from Garnacha, Cabernet-Sauvignon and Prieto Picudo grapes
which have an intermediate TPI value (TPI of 15), appeared in the middle
region. Also in the case of musts, the loading plot confirmed the
complementarity of the sensors forming the array (Figure 5.6.b).
5.4 CONCLUSIONS
A bioelectronic tongue formed by three sensors based on Langmuir-
Blodgett films of AA containing Tyr or Lac and using LuPc2 as electron
mediator was constructed. The increase in the area per molecule
observed in the surface pressure-area isotherms and the AFM images
indicated that enzymes were imbibed in the floating monolayer formed
by LuPc2 and AA with different kinetics. The hydrophobic environment
provided by the floating film was more favorable for the tyrosinase
which is a transmembrane protein.
The electrochemical responses of the sensors towards phenolic

compounds (mono, di and triphenols) depended on the nature of the
enzymes and on their different enzymatic specificity. The biomimetic
environment improved the dynamic behavior and the detection limits
which were in the range of 10-7 and 10-8 mol·L-1.
137
The bioelectronic tongue was able to discriminate phenols according to

their chemical nature. In addition, the system was able to discriminate
musts according to their Total Polyphenol Index. The loading plots
confirmed the excellent complementarity of the sensors forming the
array.
It has been demonstrated that the proposed array of sensors combines

the advantages of classical phthalocyanine based-sensors that provide
global information about the sample, with the specificity of the enzyme-
substrate reaction typical of biosensors. For this reason, the selectivity
of the multisensor system and its capability of discrimination were
clearly improved when biosensors containing Tyr and Lac were included
in the array.
The high functionality of the enzyme obtained using a biomimetic

immobilization method, the selectivity afforded by enzyme catalysis and
the signal enhancement caused by the LuPc2 mediator make this
bioelectronic tongue attractive for the analysis of grapes.
References
1. Mello, L.D., Del Río, D., Costa, L.G., Lean, M.E.J., Crozier, A. Nutr.
Metab. Cardiovasc. Dis. 2010, 20, 1-6.
2. Bartosz, G. Food oxidants and antioxidants: chemical, biological, and

functional properties (chemical & functional properties of food
components), 2013, CRC press, Florida (USA).
138
3. Vlasov, Y., Legin, A., Rudnitskaya, A., Di Natale, C., D'Amico, A. Pure
Appl. Chem. 2005, 77, 1965-1983.
4. Tahara, Y., Toko, K. IEEE Sens. J. 2013, 13, 3001-3011.
5. Ciosek, P., Wroblewski, W. Sensors. 2011, 11, 4688-4701.
6. Scampicchio, M., Ballabio, D., Arecchi, A., Cosio, S.M., Mannino, S.

7. Winquist, F., Olsson, J., Eriksson, M. Anal. Chim. Acta. 2011, 10, 192-
197.
8. Escobar, J.D., Alcaniz, M., Masot, R., Fuentes, A., Bataller, R., Soto, J.,
Barat, J.M. Food Chem. 2013, 138, 814-820.
9. Rodriguez-Mendez, M.L., Parra, V., Apetrei, C., Gay, M., Prieto, N., de
Saja, J.A. Microchim. Acta. 2008, 163, 23-31.
10. Cabral, F.P.A., Bergamo, B.B., Dantas, C.A.R., Riul, A., Giacometti, J.A.
Rev. Sci. Instrum. 2009, 80, 026107-026107-3
11. Gay, M., Muñoz, R., De Saja, J.A., Rodriguez-Mendez, M.L.

12. Ceto, X., Gutierrez, J.M., Gutierrez, M., Céspedes, F., Capdevila, J.,
Mínguez, S., Jimenez-Jorquera, C., del Valle, M. Anal. Chim. Acta. 2012,
732, 172-179.
Chem. Rev. 2010, 254, 2755–2794.
14. Bouvet, M., Gaudillat, P., Suisse, J.M. J. Porphyrins Phthalocyanines.

2013, 17, 913-919.
15. Matemadombo, F., Apetrei, C., Nyokong, T., Rodriguez-Mendez,

M.L., de Saja, J.A. Sensors. Actuat. B. 2012, 457, 166-167.
139
16. Arrieta, A., Rodriguez-Mendez, M.L., de Saja, J.A. Sensors Actuat. B.

2003, 95, 357-365.
M., Riul, A., Oliveira, O.N.Jr., Constantino, C.J.L. J. Phys. Chem. B. 2008,
112, 15275-15282.
18. Karim, F., Fakhruddin, A. N. M. Rev. Environ. Sci. Biotechnol. 2012,

11, 261-274.
19 Yin, H.S., Zhou, Y.L., Xu, J., Ai, S.Y., Cui, L., Zhu, L.S. Anal.Chim.Acta.
2010, 659, 144-150.

J.A., Zucolotto, V. Biosens Bioelectron. 2011, 26, 4715-4719.
21. Apetrei, C., Alessio, P., Constantino, C.J.L., De Saja, J.A., Rodriguez-
Mendez, M.L., Pavinatto, F.J., Fernandes, E.G., Zucolotto, V., Oliveira,
O.N. Jr. Biosens. Bioelectron. 2011, 26, 2513-2519.
22. Soloducho, J., Cabaj, J., Swist, A. Sensors. 2009, 9, 7733-7752.
23. Ceto, X., Cespedes, F., Pividori, M.I., Gutierrez, J.M., del Valle, M.
Analyst. 2012, 137, 349-356.
24. Ghasemi-Varnamkhasti, M., Rodriguez-Mendez, M.L., Mohtasebi,

S.S., Apetrei, C., Lozano, J., Ahmadi, H., Razavi, S.H., de Saja, J.A. Food
Control. 2012, 25, 216-224.
25. Linaje, M., Quintanilla, M.C., Gonzalez, A., del Valle, J.L., Alcaide, G.,
Rodriguez- Mendez, M.L. The Analyst. 2000, 125, 341-346.
26. Pavinatto, F.J., Fernandez, E.G.R., Alessio, P., Constantino, C.J.L., de

Saja, J.A., Zucolotto, V., Oliveira, O.N.Jr., Apetrei, C., Rodriguez-Mendez,
M.L. J. Mater. Chem. 2011, 21, 4995-5003.
27. OIV. Compendium of International Methods of Analysis of Wines and

Musts, 2013, vol.2.
140

Electrochim. Acta. 2004, 49, 5177–5185.
29. Casili, S., De Luca, M.A., Apetrei, C., Parra, V., Arrieta, A., Valli, L., Jiang,
J., Rodriguez-Mendez,M.L., de Saja J.A. App. Surf. Sci. 2005, 246, 304-313.
30. Makhotina, O., Kilmartin, P.A. Anal. Chim. Acta. 2010, 668, 155-165.
31. Cortina-Puig, M., Noguer, T., Marty, J.L., Calas-Blanchard, C.

Electrochemical Biosensors as a Tool for the Determination of Phenolic
Compounds and Antioxidant Capacity in Foods and Beverages. In
Biosensors in Food processing, safety and quality control, 2011, 257-272.
Ed. CRC Press. Boca Raton, London (UK).
32. Tang, L., Zhou, Y.Y., Zeng, G.M., Li, Z., Liu, Y.Y., Zhang, Y., Chen, G.Q.,
Yang, G.D., Lei, X.X., Wu, M.S. Analyst. 2013, 138, 3552-3560.
141
CHAPTER 6
Array of Biosensors for Discrimination of Grapes
According to Grape Variety, Vintage and Ripeness
During the ripening process is necessary to monitor not only

polyphenolic content but also sugar concentration in the berries. Sugar
content is one of the most important parameters analyzed by the
winemaking industry as it is directly related to the alcohol degree of the
wine.
There are two kinds of sugar mainly present in the grapes, D-glucose and
D-fructose. Their ratio during the ripening process varies however when
the berries are ripe the ratio is close to the unity.
Due to the excellent performance of the bioelectronic tongue containing

phenol oxidases we decided to incorporate to the array new enzymes
devoted to the analysis of these two molecules (glucose oxidase and D-
fructose dehydrogenase). The methodology needed to be adapted in
order to deposit the films containing the new enzymes. This
bioelectronic tongue has been applied to discriminate musts prepared
from different varieties of grapes, grapes harvested in two different
vintages and to monitor the ripening of grapes from veraison to harvest.
In this chapter, the multisensory system formed by four LB films

containing two phenol oxidases (tyrosinase and laccase) and two
biosensors dedicated to the detection of sugars (glucose oxidase and D-
fructose dehydrogenase) have been developed. The capability of
discrimination of the array will be evaluated using Principal Component
Analysis (PCA). The correlation of the responses with the phenolic and
sugar content of the grapes analyzed by classical chemical methods will
be evaluated using Partial Least Squares (PLS-1). This work has been
submitted to Analytica Chimica Acta.
6.1 INTRODUCTION
The quality of the grapes is typically established on the basis of their
phenolic and sugar content1,2. Sugar content is the parameter usually
used to monitor the maturity of grapes because it is directly associated
with the alcohol degree of the final product. Phenols are a large group of
metabolites with a direct influence on the organoleptic characteristics of
145
wine3. There is an increasing interest to detect the phenolic maturity of

grapes, a parameter that would be crucial to decide when to harvest.
Sugar and phenolic content of wines and musts are usually analyzed
using optical, spectroscopic or electrochemical methods4.
Electrochemical biosensors can be an alternative to classical methods
due to their high specificity, low detection limits, rapid response and
possibility of miniaturization5,6.
A large variety of biosensors dedicated to the detection and

quantification of phenols has been developed. Those sensors contain
phenol oxidases such as tyrosinase, laccase or horseradish peroxidase7.
Similarly, many efforts have been carried out to develop biosensors
devoted to the detection of sugars using glucose oxidase8,9. Biosensors
containing other enzymes such as fructose dehydrogenase have been
studied in lesser extent10,11.
The immobilization of the enzyme onto the substrate is a key step in the
development of efficient biosensors with high enzymatic activity12.
Classical immobilization strategies include adsorption on a substrate13,
immobilization in a carbon paste or PVC matrix14,15, entrapment in a
polymeric matrix16, covalent binding17, cross-linking18 or encapsulation
on liposomes among many others19. The advances in Nanoscience and
Nanotechnology provide alternative methods to immobilize
biomolecules preserving their activity more efficiently. Layer by layer
(LbL)20-22 and Langmuir-Blodgett (LB)23-25 are two examples of these
techniques. The advantages of LbL and LB films are directly associated
with the control of the thickness and the possibility of tuning the
molecular architectures. Moreover, these techniques allow preparing
sensors where the enzymes are adsorbed in biomimetic lipidic layers
formed by fatty acids or phospholipids (similar to membrane cells) via
COOH group interaction. This biomimetic environment contributes to
the preservation of the enzymatic conformation. This strategy has been
successfully used to prepare glucose oxidase26 and tyrosinase
biosensors27. LB technique can also be advantageous because the
146
electron mediator can be co-immobilized with the enzyme in the lipidic

layer, facilitating the electron transfer process27-29.
During the last years, multisensory systems combined with appropriate

pattern recognition software (electronic tongues) have been developed
and applied to analyze complex liquids30,31. They have been successfully
used for the discrimination and classification of wines according to their
grape variety, type of ageing or adulteration32-34 and grapes35.
It has been claimed that arrays formed by biosensors can be

advantageous because they combine the advantages of classical arrays
of electrochemical sensors (that provide global information about the
sample) with the specificity of the enzyme-substrate reaction typical of
biosensors. In spite of their good capability to discriminate wines or
beers, bioelectronic tongues have rarely been used to obtain specific
information about the composition of the samples14,36,37.
6.2.1- Chemicals
Laccase (Lac), from Trametes versicolor (EC Number: 1.10.3.2, activity of
plus of 20.7 U·mg-1), Tyrosinase (Tyr) from Mushroom (EC Number 1
1.14.18.1, activity of 3610 U·mg-1), Glucose oxidase (GOx) from
Aspergillus Niger, type VII (EC Number: 1.1.3.4, activity plus 100000
U·mg-1) and D-Fructose Dehydrogenase (FDH) from Gluconobacter
Industrius (EC Number:1.1.99.11, activity of 400-1,200 U·mg-1,) were
purchased from Sigma Chemical Co. (USA). 70 g·mL-1 solutions of the
enzymes were prepared in phosphate buffer 0.01 mol·L-1 (pH=7.0) (PBS).
All chemical and solvents (Sigma Chemical Ltd.) were of reagent grade
and used as supplied. Deionized water was obtained from a Millipore
purifier (18.2 M·cm-1).
The lutetium (III) bisphthalocyaninate (LuPc2) was synthesized and

purified in the neutral radical state following a previously published
procedure38.
147
6.2.2- Grape samples

Five varieties of grapes (Tempranillo, Garnacha, Cabernet-Sauvignon,
Prieto Picudo and Mencía) were selected and harvested in Castilla y
León region (Spain). Grapes were harvested from the same cultivar in
two consecutive vintages (2012 and 2013). In addition, during the year
2013, grapes of the variety Tempranillo were collected periodically in
order to monitor the ripening. Sample 1 corresponded to grapes
collected during the veraison (change of the grape color from green to
red). Then, samples were collected on a weekly basis (Samples 2 to 5).
To obtain the musts, 300 grams of each type of grape were crushed
using standard procedures. Chemical analyses were carried out
following international regulations4. Parameters measured included
classical chemical markers of the sugar content (degree brix, density,
sugars, grade (16.8), grade (17.5)), markers of the polyphenolic content
(Total Polyphenol Index, TPI) and indicators of the acidity (pH, total
acidity, malic acid, tartaric acid).
6.2.3- Langmuir-Blodgett films

LB sensors containing enzyme (Enz), fatty acid (arachidic acid, AA) and
the electron mediator (lutetium bisphthalocyanine, LuPc2)
(Enz/AA/LuPc2), were fabricated using a KSV 2000 trough (KSV
Instruments, Finland). Films of Tyr/AA/LuPc2 and Lac/AA/LuPc2 were
prepared using the conditions previously described33. LB film sensors
based on FDH and GOx are reported here for the first time and
preparation conditions were established in order to maximize the
insertion of the enzyme into the floating film.
GOx and FDH films were prepared at pH 4.5 using a phosphate buffered
saline (PBS-NaCl) subphase (NaCl 0.1mol·L-1, phosphate buffer 0.01 mol·
L-1, pH=4.5). The Langmuir monolayers at the air/water interface were
characterized by surface pressure versus mean molecular area (π-A)
isotherms at 21C by spreading 100 μl of a chloroform AA/LuPc2 mixture
(10:1) onto the subphase. Once the solvent was evaporated, 40 μl of a
70 μg·ml-1 solution of the corresponding enzyme (GOx or FDH) in 0.01
mol·L-1 PBS (pH=4.5) were injected drop by drop underneath the
148
air/liquid interface. The enzymes were then adsorbed inside the floating
film for one hour. Barriers were compressed at a speed of 5 mm·min-1.
Enz/AA/LuPc2 transferred to ITO glass at a surface pressure of 40 mN·m-1

by Y-type deposition with a transfer ratio close to 1 with a substrate
speed of 3 mm·min-1. Finally, the Enz/AA/LuPc2 LB films were treated
with glutaraldehyde (2.5% w/v in PBS).
Brewster Angle Microscope (BAM) images were registered using a

KSV/Nima MicroBAM

Biosensors were used as working electrode, the reference electrode was
Ag|AgCl/KCl 3 M and the counter electrode was a platinum plate. The
potentiostat was an EG&G PARSTAT 2273 potentiostat/galvanostat.
Cyclic voltammograms were registered at a sweep rate of 0.1 V·s-1 from
-0.8 V to +1.0 V. In order to facilitate the diffusion of ions and to reduce
the complexity of the sample, musts were diluted 50% in water.
6.2.5- Statistical analysis.

Curves were pre-processed a data reduction technique based on
“kernels”39. The voltammogram curve is multiplied by 10 smooth, bell-
shaped windowing functions defined as Equation 6.1:
1
𝐾𝑖 (𝑉𝑗 ) = 𝑉𝑗−𝑐𝑖 (Equation 6.1)
1+( )2𝑏𝑖
𝑎𝑖
Where ai, bi and ci define the width, shape and center of the different
windowing functions Ki. Subsequently, data were integrated with
respect to voltage. After compression, each voltammogram has been
reduced to a vector of 10 variables which were used as input data
source for statistical analysis.
A non-supervised multivariate method, Principal Component Analysis

(PCA), was used to evaluate the capability of discrimination of the
system. Partial Least Squares (PLS-1) method was used to establish
149
correlations between the signals produced by the array of biosensors

and the results obtained by means of chemical methods. Chemometric
analyses were carried out using the software Matlab v5.3. (The
Mathworks Inc., Natick, MA, USA).
6.3.1- Characterization of Langmuir films

The first step was to develop a method to prepare GOx/AA/LuPc2 and
FDH/AA/LuPc2 LB films. For this purpose, Surface pressure/Area (/A)
isotherms of the Enz/AA/LuPc2 systems were registered. Figure 6.1
shows the isotherms registered for the four sensing layers included in
the array. The isotherm obtained from the mixture of the lipid and the
electron mediator (AA/LuPc2) has been included in the graph for
comparison purposes.
The limiting area obtained from the mixture AA/LuPc2 was 37 Å2. As
expected, the incorporation of enzymes produced the expansion of the
isotherms to higher areas per molecule
The limiting area per molecule values obtained for GOx/AA/LuPc2 and
FDH/AA/LuPc2, with values of 63 mN·m-1 and 60 mN·m-1 respectively,
were clearly higher than those observed in phenol oxidases (47 mN·m-1
for Lac/AA/LuPc2 and 51 mN·m-1 for Tyr/AA/LuPc2). This indicated that
enzymes were adsorbed in the amphiphilic monolayer. The large values
observed in GOx and FDH can be due to an enhanced adsorption of the
enzyme into the AA environment caused by more favorable
hydrophobic-hydrophilic interactions between the enzymes and the
floating AA/LuPc2 film. The differences in molecular weight of the
enzymes must also have an important effect (120-133 KDa for Tyr, 50-
100 KDa for Lac, 160 KDa for GOx, 150 KDa for FDH). In fact, GOx and
FDH which present the higher molecular weights also showed the larger
areas per molecule.
In order to better understand the intermolecular interactions inside the

floating films, compression-expansion cycles were analyzed (Figure 6.1.b
150
and 6.1.c). Under these conditions, the behavior of GOx and FDH
enzymes was different from the behavior shown by phenol oxidases
which are desorbed during the expansion cycle40. In the case of
GOx/AA/LuPc2 and FDH/AA/LuPc2 the limiting area per molecule
obtained after decompression was only slightly smaller than the value
observed in the first compression. During the second compression, the
area per molecule values were only slightly smaller than those
registered in the first compression. Therefore, only a small amount of
GOx or FDH was desorbed from the film during the expansion. In
addition, a large part of this desorbed enzyme was readsorbed during
the second compression. Because GOx and FDH were not completely
expelled from the film during decompression and were easily
readsorbed, it can be concluded that the interactions with the AA/LuPc2
film were stronger than in the case of Tyr and Lac, justifying the larger
area per molecule observed.
The homogeneity of the floating films was analyzed using Brewster

Angle Microscopy (BAM) (Figure 6.1.b and 6.1.c inset). Good quality
surfaces were observed for all the films analyzed. BAM images also
showed a clear increase in the roughness of the film (with respect to the
AA/LuPc2 film). This increase was particularly evident in GOx/AA/LuPc2
films that could be associated with the high affinity of the enzyme
towards the lipidic layer.
151
Figure 6.1. a)Pressure area isotherms registered in PBS-NaCl subphase for

AA/LuPc2 and Enz/AA/LuPc2 films where Enz is Tyrosinase (Tyr), Laccase (Lac),
Glucose oxidase (GOx) and D-Fructose dehydrogenase (FDH). b) Compression-
expansion cycles of GOx/AA/LuPc2 and c) Compression-expansion cycles of
FDH/AA/LuPc2. First cycle (red line), second cycle (blue line). Inset show BAM
images.
6.3.2- Discrimination of musts obtained from different varieties

of grapes harvested in two consecutive vintages
The array formed by four biosensors was used to analyze musts
prepared from five varieties of grapes (Tempranillo, Garnacha,
Cabernet-Sauvignon, Prieto Picudo and Mencía) harvested in two
different vintages (2012 and 2103) from the same cultivar.
The response of the array is illustrated in Figure 6.2 for Prieto Picudo
variety. The peaks related to the antioxidants appeared as anodic waves
in the 0.5-1.0 V region with the corresponding cathodic peaks at ca. 0.6
V. The peaks associated with sugars appeared at negative potentials (ca.
-0.2 V). Due to the complexity of the media, the peaks were wider than
those observed in standard aqueous solutions41. Each electrode
provided a different response towards the same must sample and the
152
positions and the intensity of the peaks changed with the grape variety.
Moreover, voltammograms could detect small differences from one
vintage to another; however, the general behavior of grapes of the same
variety was maintained.
Figure 6.2. Voltammetric responses of the four sensors towards must prepared
from the grape variety Prieto Picudo harvested in 2012. Glucose oxidase (pink
line), Laccase (red line), D-Fructose dehydrogenase (green line) and Tyrosinase
(blue line).
When PCA was carried out in musts obtained from grapes of different
varieties (harvested in 2012), the scores plot showed that the array was
able to discriminate musts according to the type of grape. The three first
components explained ca. 97% of the total information of the system.
The experiment was repeated in the vintage 2013 and the results were
similar. Moreover, PCA was carried out merging in the same matrix the
responses obtained in both 2012 and 2013 (Figure 6.3). Grapes were
discriminated according to their variety. Samples collected in different
vintages could be clearly distinguished but appeared close one to each
other in the scores plot. This result indicates that the bioelectronic
tongue provides similar results than classical chemical analyses that
clearly establish that varietal qualities are maintained from one year to
another.
153
Figure 6.3. a) PCA scores plot and b) loading plot illustrating the discrimination
capability of the bioelectronic tongue exposed to must prepared from the same
five varieties harvested in two vintages (2012 and 2013): 1-Cabernet-Sauvignon
2012 (purple), 2-Cabernet-Sauvignon 2013 (yellow), 3-Garnacha 2012 (pink), 4-
Garnacha 2013 (light blue), 5-Mencía 2012 (dark blue), 6-Mencía 2013 (red), 7-
Tempranillo 2012 (black), 8-Tempranillo 2013 (brown), 9-Prieto Picudo 2012
(green), 10-Prieto Picudo 2013 (orange).
The position of the clusters was directly related to the chemical

composition of the grapes. For this reason, the must prepared from
Mencía variety, with the lowest concentration of sugar (average of 215
g·L-1) and TPI (average of 12), was located in the left side of the diagram
whereas, samples from Tempranillo variety with the highest sugar
content (average of 238.2 g·L-1) and TPI (average of 20) were located in
the top part of the figure. Musts obtained from Garnacha, Cabernet-
Sauvignon and Prieto Picudo grapes with intermediate values of sugar
(220-230 g·L-1) and TPI (TPI of 15), appeared far apart from the rest. pH
seems to play an important role in the response of the array:
Tempranillo and Mencía samples showed the highest pHs (4.18 and 4.17
respectively) and were located far from Prieto Picudo (3.7) and
Cabernet-Sauvignon (3.8). Finally, Garnacha which showed the lowest
pH (3.4) was clearly discriminated from the rest.
Figure 6.3.b shows the contribution of the variables (10 kernels per
sensor) in the PCA loading plot. The loadings of each sensor were
located in a different quadrant and showed values close to 1. This is
important because each one of the four sensors forming the array
154
contributes in an important manner to the discrimination process. This

also evidences the high cross-selectivity attained by the four biosensors.
6.3.3- Monitoring the grape ripeness by means of the

bioelectronic tongue
The bioelectronic tongue was used to analyze grapes harvested from the
veraison to ripening in a weekly basis during five weeks (variety
Tempranillo, vintage 2013). Sample 1 was collected during the veraison
(change of the grape color from green to red). Samples 2-5 were
harvested in subsequent weeks. Sample number 4 corresponded to the
optimal date of harvesting predicted by chemical parameters. Sample 5
corresponded to over-ripened grapes.
Figure 6.4. Response of the Tyr/AA/LuPc2 towards all the samples of

Tempranillo harvested weekly from veraison to over-ripeness (vintage 2013).
First week (dark blue line), second week (orange line), third week (green line),
fourth week (light blue line) and fifth week (pink line).
Figure 6.4 illustrates how voltammograms changed with the ripening

using a tyrosinase electrode. The anodic peak at ca. 0.7 V associated
with polyphenols shifted to higher potentials whereas the peaks at ca.
0.4 V and -0.1 V increased their intensity progressively from weeks 1 to
3. Suddenly, in week 4 (the optimal date as indicated by classical
techniques) the intensity decreased and was maintained in over-ripened
grapes collected in week 5. These changes are in good agreement with
155
changes in the phenolic maturation measured by chemical methods

which indicate that the development of phenolic maturation in the late
stage of ripening produces a notorious increase near the harvest42.
Similar results were obtained with biosensors modified with laccase. In
turn, sensors modified with glucose oxidase could follow the increase in
the sugar content by showing an increase in the intensity of the cathodic
peak at -0.2V.
PCA showed a clear discrimination of the samples collected during the

progression of maturing. (Figure 6.5). Clusters corresponding to samples
harvested in successive weeks from veraison to ripening (samples 1 to 4)
were located in a clockwise order. Grapes collected one week after the
optimal maturity (sample 5), did not follow the same trend than the
others.
Figure 6.5. PCA scores plot illustrating the discrimination capability of the
bioelectronic tongue exposed to must prepared from Tempranillo samples
harvested in 2013 weekly from 1-first week (veraison), 2-second week, 3-third
week, 4-fourth week (harvest), 5-fith week (over-ripe).
6.3.4. Multiparametric correlations between the bioelectronic

tongue and the classical chemical analysis
Table 6.1 shows the statistical parameters for the PLS-1 regression
models correlating the output of the array with the chemical
parameters.
156
a
Table 6.1. Results of the PLS-1 analysis. Root Mean Square Error of Calibration;
b c
Root Mean Square Error of Prediction. Square correlation coefficient in
d e
calibration. Square correlation coefficient in prediction. Latent variables.
Parameter RMSECa RMESPb RCc RPd LVe

Degree brix 0.254 0.307 0.987 0.982 3
Density 0.003 0.005 0.904 0.732 2
Sugars 0.801 1.298 0.902 0.729 2
Grade (16.8) 0.475 0.770 0.902 0.729 2
Grade (17.5) 0.458 0.738 0.900 0.725 2
pH 0.058 0.070 0.982 0.974 3
Total acidity 0.047 0.059 0.995 0.992 3
Malic acid 0.209 0.246 0.927 0.897 3
Tartaric acid 0.126 0.143 0.996 0.994 3
TPI 0.593 0.677 0.973 0.965 3
Excellent correlations were found for all the parameters related to the
sugar (sugar, degree brix, grade and density) and in particular with the
degree Brix which is commonly used to evaluate the quality of grapes.
Also good correlations were found with TPI. These good correlations can
be attributed to the excellent performance of the enzymes imbibed in
the LB films. It is also worthy to notice the good correlations found with
parameters related to the pH (pH, Total acidity, Malic and Tartaric acid).
This can be explained by the influence of the pH in the enzymatic
activity.
6.4. CONCLUSIONS
A bioelectronic tongue formed by four voltammetric biosensors based
on LB films containing GOx, FDH, Tyr and Lac, LuPc2 as electron mediator
and AA as lipid, was developed. The increase in the area per molecule
observed in the surface pressure-area isotherms confirms that enzymes
were imbibed inside the floating monolayer formed by LuPc2 and AA.
The array of biosensors was able to discriminate five varieties of grapes

according to their sugar and phenolic composition. The system was
more sensitive to the differences caused by the variety of grape than to
157
the changes caused by weather conditions from vintage to vintage. It

was also able to monitor the ripening of grapes from veraison to
harvesting and could be used to predict the appropriate date to harvest.
The loading plot confirmed the excellent complementarity of the

biosensors. Moreover, PLS-1 showed exceptional correlations with all
the chemical parameters directly related to phenols and sugar
concentration.
158
References
1. Boulton, R. Am. J. Enol. Vitic. 2001, 52, 67-87.
2. Glories, Y. Progrès Agricole et Viticole. 2001, 118, 347-350.
3. Sun, B.S., Pinto, T., Leandro, M.C., Ricardo-Da-Silva, J.M., Spranger,

M.I. Am. J. Enol. Vitic. 1999, 50, 179-184.
4. OIV, Compendium of international methods of analysis of wines and

musts. Vol. 2, 2013.
5. Sassolas, A., Blum, L.J., Leca-Bouvier, B.D. Biotechnol. Adv. 2012, 30,
489-511.
6. Thevenot, D.R., Toth, K., Durst, R.A., Wilson, G.S. Biosens. Bioelectron.
2001, 16, 121-31.
7. Fahmida, K., Fakhruddin, A.N.M. Rev. Environ. Sci. Biotech. 2012, 11,
261-274.
8. Nouira, W., Maaref, A., Elaissari, H., Vocanson, F., Siadat, M., Jaffrezic-
Renault, N. Mater. Sci. Eng. C. 2013, 33, 298-303.
9. Kamyabi, M.A., Hajari, N., Turner, A.P.F., Tiwari, A. Talanta. 2013, 116,
801-808.
10. Antiochina, R., Vinci, G., Gorton, L. Food Chem. 2013, 140, 742-747.
11. Damar, K., Demirkol, D.O. Talanta. 2011, 87, 67-73.
12. Goto, T.E., Lopez, R.F., Oliveira, O.N., Caseli, L. Langmuir. 2010, 26,
11135-11139.
13. Moyo, M., Okonkwo, J. Sensors Actuat. B. 2014, 193, 515-521.
14. Cetó, X, Capdevilla, J., Mínguez, S., del Valle, M. Food Res. Int. 2014,
55, 455-461.
159
15. Apetrei, I.M., Rodriguez-Mendez, M.L., Apetrei, C., de Saja, J.A.

Sensors Actuat. B. 2013, 177, 138-144.
16. Mariani, A.M., Natoli, M.E., Kofinas, P. Biotechnol. Bioeng. 2013, 10,
2994-3002.
17. Dong, J., Liu, T., Meng, X., Zhu, J., Hang, K., Ai, S., Cui, S. J. Solid State
Electrochem. 2012, 16, 3783-3790.
18. Vicentini, F. C., Janegitz, B. C., Brett, C. M. A., Fatibello-Filho, O.

Sensors Actuat. B. 2013, 188, 1101-1108.
19. Guan, H., Liu, X., Wang, W. J. Solid State Electrochem. 2013, 17,
2887-2893.
20. Ho, M.L., Wang, J.C., Wang, T.Y., Lin, C.Y., Zhu, J.F., Chen, Y.A., Chen,
T.C. Sens. Actuators B. 2014, 190, 479-485.
21. dos Santos, V., dos Santos, M., de Jesus, C.G., Fujiwara, S.T., Garcia,
J.R., Pessoa, C.A., Wohrath, K. Int. J. Electrochem. Sci. 2013, 8, 10601-
10620.

23. Cabaj, J., Soloducho, J., Nowakowska-Oleksy, A. Sensors Actuat. B.

2010, 143, 508-515.
24. Zanon, N.C.M., Oliveira Jr, O.N., Caseli, L. J. Colloid Interface Sci.
2012, 373, 69-74.
25. Challier, L., Gal, F., Carrot, G., Perez, H., Noel, V. Electrochem.
Commun. 2013, 28, 118-121.
26. Wang, K.H., Syu, M.J., Chang, C.H., Lee, Y.L. Sensors Actuat. B. 2012,
164, 29-36.
160
27. Pavinatto, F.J., Fernandes, E.G.R., Alessio, P., Constantino, C.J.L., de

Saja, J.A., Zucolotto, V., Apetrei, C., Oliveira Jr, O.N., Rodriguez-Mendez,
M.L. J. Mat. Chem. 2011, 21, 4995-5003.
28. Alessio, P., Pavinatto, F.J., Oliveira Jr, O.N., de Saja, J.A., Constantino,
C.J.L., Rodriguez-Mendez, M.L. Analyst. 2010, 135, 2591-2599.
29. Zagal, J.H., Griveau, S., Ozoemena, K.I., Nyokong, T., Bedioui, F.
Nanosci. Nanotechnol. J. 2009, 9, 2201-2214.
30. Sliwinska, M., Wisniewska, P., Dymerski, T., Namiesnik, J., Wardenci,
W. J. Agric. Food. Chem. 2014, 62, 1423-1448.
31. Smyth, H., Cozzoino, D. Chem. Rev. 2013, 113, 1429-1440.
32. Gay, M., Apetrei, C., Nevares, I., del Alamo, M., Zurro, J., Prieto, N.,
de Saja, J.A., Rodriguez-Mendez, M.L. Electrochim. Acta. 2010, 55, 6782-
6788.
33. Apetrei, C., Apetrei, I.M., Nevares, I., del Alamo, M., Parra, V.,
2594.
34. Prieto, N., Gay, M., Vidal, S., Aagaard, O., de Saja, J.A., Rodriguez-
Mendez, M.L. Food Chem. 2011, 129, 589-594.
35. Moreno-Codinachs, L., Kloock, J.P., Schöning, M.J., Baldi, A., Ipatov,
A., Bratov, A., Jiménez-Jorquera, C. Analyst. 2008, 133, 1440–1448.
36. Ghasemi-Varnamkhasti, M., Rodriguez-Mendez, M.L., Mohtasebi,

S.S., Apetrei, C., Lozano, J., Ahmadi, H., Razavi, S.H., de Saja, J.A. Food
Control. 2012, 25, 216-224.
37. Medina-Plaza, C., Revilla, G., Munoz, R., Fernandez-Escudero, J.A.,

Barajas, E., Medrano, G., de Saja, J.A., Rodriguez-Mendez, M.L. J.
161
38. Linaje, M., Quintanilla, M.C., del Valle, J.L., Alcaide, G., Rodriguez-
Mendez, M.L. Analyst. 2000, 125, 341-346.
39. Parra, V., Hernando, T., Rodriguez-Mendez, M.L., de Saja, J.A.

40. Medina-Plaza, C., de Saja, J.A., Rodriguez-Mendez, M.L. Biosens.

Bioelectron. 2014, 57, 276-283.
41. Zhang, H., Meng, Z., Wang, Q., Zheng, J. Sensors Actuat. B. 2011, 158,
23-27.
42. Ribéreau-Gayon, P., Glories, Y., Maujean, A., Dubourdieu, D.

Handbook of enology. Vol. 2. The chemistry of wine stabilization and
treatments. 2000. Ed. John Wiley & Sons, Chichester (England).
162
CHAPTER 7
Electrochemical Characterization of Dilithium
Phthalocyanine Carbonaceous Electrodes
Despite the excellent performance of nanostructured (bio)sensors,

classical techniques were used to obtain cheaper sensors that could be
easily applied in the industrial environment.
Carbon Paste Electrode (CPE) technique is an easy and fast method to

obtain voltammetric sensors. These sensors consist on a composite
made mixing carbonaceous powder with a binder packed into a plastic
tube using a metallic copper wire as electrical contact. One of the
advantages of these sensors is that they can be used several times (i.e.
the surface can be polished regenerating the electrode).
The modification of this kind of sensors is done by mixing the modifier

with the carbonaceous material and the binder in the desire proportion
or using casting method. Different types of carbonaceous materials can
also be used as a matrix (microspheres, graphite powder, carbon black,
nanotubes…).
In this chapter, carbon paste electrodes of dilithium phthalocyanine

(Li2Pc) have been prepared using different carbonaceous materials:
graphite, carbon microspheres and multiwall carbon nanotubes
(MWCNT). Microspheres present the advantage of their high specific
surface and porous structure whereas carbon nanotubes are excellent
candidates for the fabrication of electrochemical sensors because of
their electrical conductivity, large surface area and ability to reduce
overpotentials. The electrochemical reduction/oxidation behaviors of
the sensors in a variety of simple electrolytes will be studied in detail
analyzing the influence of the anions, cations and the pH of the solution.
The possibility of using the Li2Pc as sensitive material and their
electrocatalytic properties towards citric acid will be analyzed. The
synergistic effect of MWCNT and Li2Pc will be evaluated. The results
presented in this chapter have been published in Journal of
Phthalocyanines and Porphyrins.
7.1 INTRODUCTION
Phthalocyanines can be immobilized on inert electrode surfaces such as
graphite, carbon, ITO glass, or gold and exhibit electrocatalytic activity
165
for a variety of reactions1-4. These metal complexes act as mediators by

lowering the overpotential of oxidation or reduction of the target
molecules or increasing the intensity of the peaks observed. Therefore,
phthalocyanine sensors have been applied in amperometric,
voltammetric or potentiometric electrocatalytic determination of many
organic and inorganic compounds.
Their catalytic action can be described in terms of chemical structure

and chemical and physical properties. Their reactive centers are clearly
identified and their reactivity can be modulated by changing the nature
of the central metal or by modifying the structure of the macrocyclic
ligand2,5,6.
The number of works published using phthalocyanines as sensitive

materials for electrochemical sensors is quite large and include a variety
of phthalocyanine derivatives (e.g. central metal ions, substituents),
electrode designs, preparation techniques and target molecules5,6.
Metallophthalocyanine complexes (MPc) where the phthalocyanine ring
(in oxidation state −2) is coordinated to a range of transition metal ions
(in oxidation state +2) have been widely studied. Other families of
phthalocyanines have been studied in lesser extent. For instance, some
works have been dedicated to electrochemical sensors based on
lanthanoid bisphthalocyanines which are sandwich type compounds
with free radical character and ring oxidation state -1.57-10.
The dilithium complex (Li2Pc) is an example of a different class of

phthalocyanine derivative, where the Pc ring is coordinated to two
lithium ions11. Both cations cannot be accommodated in the center of
the cavity and thus the lithium atoms are projected out of the plane of
the phthalocyanine ring. Dilithium phthalocyanine possess mixed
electronic-ionic conductivity due to overlap of orbitals (electronic)
and mobility of Li+ ion (ionic) in a channel formed due to stacking of the
macromolecules12. The electrochemical oxidation of the dilithium
phthalocyanine, produces the lithium complex (LiPc) (ring oxidation
state −1) that can be considered the simplest representative of the
radical phthalocyanines11,13-15. As during oxidation lithium ions are
166
liberated, these molecules have been used to prepare electrodes for

solid state batteries13. Such electrodes have been prepared by
electrodeposition16 or by Langmuir-Blodgett technique12. In spite of
these interesting electrochemical and conducting properties that make
Li2Pc a good candidate as electrode modifier, electrochemical sensors
based on Li2Pc have not been yet developed.
Electrochemical sensors based on phthalocyanines can be prepared

using a variety of techniques that include polymerization17, sol-gel18,
Langmuir-Blodgett4,19 or Layer by Layer20,21 among others. One of the
classical methods to prepare electrodes based on phthalocyanines is the
carbon paste technique (CPE) in which the phthalocyanine is mixed with
a carbonaceous material and a binder to form a paste. In addition,
different types of carbonaceous materials can be used to prepare the
paste22 that include graphite, microspheres or carbon nanotubes among
others23-36.
7.2.1- Chemicals
All reagents, including Li2Pc, were of high purity and used without
further purification (Sigma–Aldrich). Solutions were prepared using
deionized water (18.3 M·cm resistivity), Milli-Q, Millipore.
Carbonaceous materials used for the preparation of the CPE electrodes
were: Graphite powder (High purity Ultracarbon®, Ultra F purity. Bay
City, MI, USA), carbon microspheres (from Sigradur G HTW,
Thierhaupten, Germany) and carbon nanotubes (multi wall nanotubes,
Nanoledge Inc., Boncherville, Quebec, Canada). Carbon paste electrodes
containing Li2Pc were prepared by mixing the carbonaceous material,
the dilithium phthalocyanine (15%, w/w) and Nujol (Fluka) as binder of
the composite mixture34. Three types of electrodes were prepared using
different carbonaceous materials: graphite powder was used to prepare
classical graphite based carbon paste electrodes (Li2Pc/G-CPE),
microspheres were used to prepare electrodes with high surface to
volume ratio (Li2Pc/ME-CPE) and multiwall carbon nanotubes were used
to prepare electrodes that combine two electrocatalysts, Li2Pc and
167
MWCNT (Li2Pc/MWCNT-CPE). Pastes were packed into the body of a

1mL PVC (polyvinyl chloride) syringe and compressed. A metallic copper
wire was used as electrical contact.
7.2.2- Equipment
Electrochemical experiments were performed using a three electrode
system with Ag|AgCl/KCl 3M reference electrode, 1 cm2 platinum
counter electrode and the corresponding CPE as working electrode. The
potentiostat used was an EG&G PARC, Model 2273
potentiostat/galvanostat (Princeton Applied Research Corp.).
Temperature control at 25C was achieved using a circulating bath
(Neslab).
7.3.1- Electrochemical characterization

Electrochemical studies were carried out in KCl 0.1 mol·L-1. Figure 7.1
shows the cyclic voltammetric (CV) curves from -1.0 V to +1.0 V at 0.1
V·s-1 for Li2Pc carbonaceous electrodes prepared using graphite (Figure
7.1.a), microspheres (Figure 7.1.b) and nanotubes (Figure 7.1.c). The
electrode prepared with graphite (Li2Pc/G-CPE) showed a quasi-
reversible redox process at E1/2= -0.04 V with a peak separation (E) ca.
0.52 V. According to literature, this redox process is assigned to
phthalocyanine ring oxidation/reduction (from Pc2- to Pc-) that gives rise
to the LiPc radical species35. The responses were highly stable and the
electrodes could be cycled up to 100 cycles without important changes
in the responses. Only a slight decrease of the intensity of the peaks was
observed (about 5% for 100 cycles).
Electrodes prepared with microspheres (Li2Pc/ME-CPE) (Figure 7.1.b)

and nanotubes (Li2Pc/MWCNT-CPE) (Figure 7.1.c) showed a similar
behavior, but molecular aggregation of phthalocyanines influenced
strongly the electrochemical responses. The use of carbon forms with
higher active surface (as microspheres) caused a decrease of the
separation between the anodic and the cathodic waves (E= 0.504V).
This decrease was clearly more marked when using MWCNT as
168
carbonaceous material (E= 0.426V). This seems to indicate that the

electrochemical oxidation of the Li2Pc was facilitated in the presence of
MWCNT due to the  interactions with the carbon nanotubes that
promoted the charge transfer through the electrode interface and/or
the charge transport within the film confirming the electrocatalytic
effect of nanotubes. In contrast, MWCNT made difficult the reduction
process, probable due to the strong interaction between the nanotubes
and the radical species formed upon oxidation (LiPc).
Figure 7.1. Cyclic voltammograms of Li2Pc CPE electrodes prepared using: a)

graphite, b) microspheres and c) MWCNT as carbonaceous material in 0.1 mol·
-1 -1
L KCl. Scan rate, 0.1 V·s
The dynamic character of the electrode process was further examined

by registering voltammograms at different scan rates (Figure 7.2). In all
three electrodes the intensity of the peaks depended linearly with the
square root of the scan rate indicating a diffusion controlled process in
accordance with the Randles-Sevcik equation (Equation 7.1).
I=2.687×105 n3/21/2 D1/2 AC (Equation 7.1)
Where I is the anodic peak current (A), n is the number of electrons

involved in the redox process,  is the scan rate (V·s-1), D is the diffusion
coefficient (cm2·s-1), A is the electrode surface area (cm2), and C is the
concentration (mol·L-1).
This Nerstian behavior was in good accordance with data obtained using
other radical phthalocyanine-CPE sensors23. Table 7.1 shows the values
of the slope of the curves obtained by representing the intensity of the
169
anodic peaks versus the square root of the scan rate. As observed in the
table, the slope of the curve obtained when using MWCNT was
approximately one order of magnitude higher than Li2Pc/G-CPE and
Li2Pc/ME-CPE electrodes, suggesting a faster response. From the slope
of I versus 1/2 plot, the diffusion coefficient D (i.e. ions diffusing
inside/outside the film to maintain the electroneutrality) could be
calculated (Table 7.1). Again, the diffusion coefficient obtained for Li2Pc
immobilized in carbon nanotubes is higher than the value obtained using
other carbonaceous forms.
2
Table 7.1. Slope (m), coefficient of correlation (R ) and diffusion coefficients (D)
obtained by representing the intensity of the anodic peak versus the square
root of the scan rate.
Electrode m (slope) R2 D (cm2·s-1)

Li2Pc/G-CPE 410-5 0.9606 8.9410-7
Li2Pc/ME-CPE 310-5 0.9936 5.0310-7
Li2Pc/MWCNT-CPE 110-4 0.9928 3.0210-6
It is worthily noting that although Li2Pc/MWCNT-CPE also showed a

Nerstian behavior, voltammograms registered at high scan rates
appeared tilted approaching progressively to an ohmic conduction.
Lithium phthalocyanine (LiPc) possesses mixed electronic-ionic
conductivity due to overlap of orbitals (electronic) and mobility of
Li+ ion (ionic) in a channel formed due to stacking of the
macromolecules12. The formation of the conductive LiPc free radical and
the liberation of Li+ ions that were trapped at the cavities of the MWCNT
electrodes can be responsible of this increase of the ohmic behavior.
Another interesting observation was that at high scan rates, the shapes
of voltammogramms registered using graphite or microspheres were
different from those obtained at low scan rates and a small second peak
at lower potentials is observed/anodic wave at ca. -0.2 V. According to
the literature, this second peak could be attributed to the oxidation of
Li2Pc to the free radical of phthalocyanine H(Pc)35. That is, from the same
starting material (the Li2Pc), two different products can be obtained
170
depending on the oxidation conditions. Usually, electrochemical

oxidation produces the radical lithium salt (LiPc), but under certain
conditions (usually chemical oxidation), the free metal HPc (where Pc is
the radical anion Pc-) can also be obtained. In our case, high scan rates
could favor this new pathway.
Figure 7.2. Cyclic voltammograms of a) Li2Pc/G-CCE, b) Li2Pc/ME-CPE and c)

-1 -1
Li2Pc/ MWCNT-CPE in 0.1 mol·L KCl at scan rates from 0.050-0.200 V·s .
In order to understand the electron transfer behavior of the chemically

modified electrodes, electrochemical measurements on Li2Pc/G-CPE,
Li2Pc/ME-CPE, Li2Pc/MWCNT-CPE in presence of ferrocyanide were
carried out. The [Fe(CN)6]3-/[Fe(CN)6]4- couple was used to characterize
the active surface (i.e. number of active sites or blocking ability of
surfaces). Results are shown in Figure 7.3.
Figure 7.3. Cyclic voltammograms of a) Li2Pc/G-CCE, b) Li2Pc/ME-CPE and c)

-3 -1 3- 4- -1
Li2Pc/MWCNT-CPE, in 10 mol·L [Fe(CN)6] /[Fe(CN)6] in 0.1 mol·L KCl. Scan
-1
rate, 0.1 V·s .
Voltammograms obtained from electrodes prepared with graphite and

microspheres showed two redox pairs associated with both Li2Pc/LiPc
171
couple (at E1/2 ca. 0.1 V) and ferrocyanide/ferricyanide couple (at ca.
0.45 V). Peaks were broad and somehow overlapped. In contrast, when
MWCNT were used, peaks were well separated and the peak related to
the Li2Pc appeared shifted to lower potentials (i.e. anodic wave that
appeared at +0.2 V in KCl shifted to -0.1 V). The most interesting feature
was that the intensity of the ferrocyanide increased drastically in the
presence of the MWCNT, indicating that the number of active sites was
amplified. The active area calculated from voltammograms registered at
different scan rates (0.050 to 1.0 V·s-1) according to the Randles-Sevcik
equation (taking into account that the diffusion coefficient for 10-3 mol·
L-1 ferrocyanide is 7.2610–6 cm2·s–1 in 1.0 mol·L-1 KCl solution) was
similar for graphite and microspheres (0.159 cm2) whereas, for the
MWCNT electrode the calculated active area was 0.25 cm2.
The voltammetric response relied on the transfer of counterions

between the electrolyte and the bulk of these materials in response to
the electron-transfer processes associated with their oxidation and
reduction. In order to evaluate the influence of the electrolyte in the
electrochemical comportment, carbonaceous electrodes were also
immersed in solutions containing different electrolytes.
7.3.2- Analysis of simple electrolytes

The role of protons was studied by immersing the electrodes in HCl. The
cyclic voltammogram of the Li2Pc/MWCNT-CPE electrode in 10-4 mol·L-1
HCl (pH = 3) is shown in Figure 7.4. In the presence of protons, the three
electrodes showed two anodic waves of similar intensity at ca. -0.2 V
and +0.25 V. The cathodic wave contained a small peak at +0.05 V and a
broad and intense peak at -0.6 V. These two redox processes could be
due to the formation of a protonated species in HCl that has also been
described in other radical species9. This protonated ring can be oxidized
at lower potentials than the neutral Li2Pc.
172
-1
Figure 7.4. Cyclic voltammograms of Li2Pc/ME-CPE in 0.1 mol·L HCl. Scan rate,
-1
0.1 V·s .
As previously established, the electrochemical oxidation of the dilithium

salt of phthalocyanine produces de monolithium radical derivative LiPc,
(that possess mixed electronic-ionic conductivity) with liberation of Li+
ions17. Due to the important role of Li+ ion in the ionic conductivity of
Li2Pc, the electrochemical behavior of these electrodes was tested in a
LiClO4 solution. As shown in Figure 7.5, the presence of Li+ ions produced
two effects in the voltammograms. The first effect was an increase in the
separation between the anodic and cathodic waves of the redox pair
Li2Pc/LiPc while voltammograms became tilted and showed an increased
ohmic behavior. This might be the result of an increase in their
conductivity through the easier transfer of the Li+ ions in the presence of
lithium in the solution. In addition, the IR drop for electrolyte motion in
carbon pores affected the double-layer formation mechanism, in which
the charge stored is recognized to be distributed. The second effect was
that the intensity of the peaks decreased upon successive cycling,
indicating that the active sites are progressively blocked. This decrease
was dependent on the surface area of the carbonaceous materials and
followed the trend Li2Pc/G-CPE>Li2Pc/ME-CPE>Li2Pc/MWCNT-CPE. That
is, in MWCNT this effect was almost negligible.
173
Figure 7.5. Cyclic voltammograms of Li2Pc CPE electrodes prepared using: a)

graphite, b) microspheres and c) MWCNT as carbonaceous material in 0.1 mol·
-1 -1
L LiClO4. Scan rate, 0.1 V·s .
When using NH4OH as electrolyte, the responses were highly similar to

those obtained in LiClO4. The cyclic voltammograms also became tilted
and pointed at each end and the intensity changed almost linearly with
potential over a very wide range (Figure 7.6). But in this case, all
modified electrodes showed improved stability and the blocking effect
was almost unexistent. These results indicate that NH4+ ions can also
participate in the ions motion inside the electrodes and to the double
layer formation, but the basic environment prevents the blocking of the
active sites.
-1
Figure 7.6. Cyclic voltammograms of Li2Pc/ME-CPE electrodes in 0.1 mol·L
-1
NH4OH. Scan rate, 0.1 V·s .
7.3.3- Analysis of citric acid

Finally, the response of the electrodes towards an antioxidant of interest
in the food industry such as citric acid was tested (Figure 7.7).
Voltammograms showed peaks of two different origins: peaks
associated with the oxidation-reduction of the citric acid present in the
174
solution and transient responses related to the electrode material

(Li2Pc). But, the important issue is that the interactions that occur
between the electrode and the citric acid can improve the
electrochemical signals.
In all three electrodes, the electrocatalytic behavior of Li2Pc caused a

decrease in the oxidation potential of the citric acid that using a bare
electrode appeared at 0.62 V, but shifted to lower potentials in the
presence of Li2Pc (0.57 V in Li2Pc/G-CPE, 0.50 V in Li2Pc/ME-CPE and 0.45
V in Li2Pc/MWCNT-CPE). In addition, the anodic peak associated with the
Li2Pc also appeared at lower potentials (-0.07 V).
Figure 7.7. Cyclic voltammograms of a) bare MWCNT-CPE and b) Li2Pc/MWCNT-

CPE electrodes in 0.4  10 mol·L citric acid. Scan rate, 0.1 V·s .
-4 -1 -1
Catalysis was also evidenced by the enhancement of the oxidation peak

current, (~10 fold compared to the bare MWCNT electrodes). The charge
passed under the curve was larger for MWCNT than for microspheres or
graphite. This result indicates that carbon nanotubes facilitated the
charge transfer through the electrode interface and/or the charge
transport within the film. The strong electrocatalytic effect observed in
the Li2Pc/MWCNT-CPE immersed in citric acid can be attributed to the
synergistic effect between MWCNT and Li2Pc that improves
electrocatalysis for the detection of citric acid. This is illustrated in Figure
175
7.7 where the response of a bare MWCNT-CPE and a Li2Pc/MWCNT-CPE

towards citric acid are shown.
7.4 CONCLUSIONS
The electrochemical behavior of carbonaceous electrodes of dilithium
phthalocyanine prepared using graphite, carbon microspheres and
multiwall carbon nanotubes, was analyzed. The three carbonaceous
materials were adequate for the immobilization of Li2Pc and showed
distinct electrochemical behaviors that were related to the nature of the
carbon material. The electrode formed by combining MWCNT and the
Li2Pc showed a higher diffusion coefficient and an increased number of
active sites. The electrochemical response registered in different
electrolytic solutions was found to be largely dependent on the nature
of the ions present in the solution. In the presence of protons, the
electrodes showed two anodic waves produced by the protonation of
the Pc ring, whereas the presence of LiClO4 or NH4OH promoted the
mobility of lithium ions increasing the ohmic conduction. The electrodes
were also used to detect citric acid, an antioxidant of interest in the food
industry. It was demonstrated that the electrocatalytic properties of the
Li2Pc facilitate the oxidation of the citric acid. The simultaneous use of
two electrocatalysts MWCNT and Li2Pc produce a synergic effect that
improves the electrocatalytic effect.
176
References
1. Jiang, J. Functional Phthalocyanine Molecular Materials. Series:
Structure and Bonding 135, 2010. Ed. Springer, Berlin (Germany).
Chem. Rev. 2010, 254, 2755-2791.
3. Rodriguez-Mendez M.L. Sensing properties of Phthalocyanines in

Encyclopedia of Sensors. 2006, 111-134. Ed. American Scientific
Publishers, California (USA).
4. Valli, L. Adv. Col. Interf. Sci. 2005, 116, 13-44.
5. Lever A.B.P. J. Porphyrins Phthalocyanines. 1999, 3, 488-499.
6. Bedioui, F., Griveau, S., Nyokong, T., Appleby, A.J., Caro, C.A., Gulppi,
M., Ochoa, G., Zagal, J.H. Phys. Chem. Chem. Phys. 2007, 9, 3383-3396.
7. Weiss, R., Fischer, J. Lanthanide phthalocyanine complexes in The

Porphyrin Handbook, 2003, 171-246. Ed. Academic Press, New York
(USA).
8. Rodríguez-Méndez, M.L., Gay, M., de Saja J.A. J. Porphyrins

9. Gay, M., Muñoz, R., de Saja, J.A., Rodriguez-Mendez, M.L. Electrochim.

Acta. 2012, 68, 88-94.
10. Parra, V., Hernando, T., Rodríguez-Méndez, M.L. de Saja, J.A.

11. Bouvet, M. Radical Phthalocyanines and Intrinsic Semiconduction in

Porphyrin Handbook. 2003, 37-104. Ed. Academic Press, New York
(USA).
12. Xiang, H.Q., Tanaka, K., Takahara, A., Kajiyama, T. Langmuir. 2002,
18, 2223–2228.
177
13. Scanlon, L.G., Lucente, L.R., Feld, W.A., Sandi, G., Balbuena, P.B.,
Alonso, P.R., Turner, A. J. Electrochem. Soc. 2004, 151, A1338-A1343.
14. Deng, X., Porter III, W.W., Vaid, T.P. Polyhedron. 2005, 24, 3004–
3011.
15. Turek, P., Petit, P., Andre, J.J., Simon, J., Even, R., Boudjema, B.,
Guillaud, G., Maitrot, M. J. Am. Chem. Soc. 1987, 109, 5119-5122.
16. Ilangovan, G., Zweier, J.L., Kuppusamy, P. J. Phys. Chem. 2000, 104,
4047-4059.
17. Griveau, S., Gulppi, M., Pavez, J., Zagal, J.H., Bedioui, F. Electroanal.
2003, 15, 779-785.
18. Guo, Y., Guadalupe, A.R. Sensors Actuat. B. 1998, 46, 213-219.
19. Gorbunova, Y., Rodríguez-Méndez, M.L., Kalashnikova, I.P., Tomilova,

L., de Saja, J.A. Langmuir. 2001, 17, 5004-5010.
20. Demel, J., Lang, K. Eur. J. Inorg. Chem. 2012, 32, 5154-5164.
M., Riul Jr., A., Oliveira, O.N.Jr., Constantino C.J.L. J. Phys. Chem. B. 2008,
112, 15275-15282.
22. Svancara, I., Vytras, K., Barek, J., Zima, J. Crit. Rev. Anal. Chem. 2001,
31, 311-345.
23. Apetrei, C., Apetrei, I.M., de Saja, J.A., Rodriguez-Méndez, M.L.

Sensors. 2011, 11, 1328-1344.
24.Ruiz-Morales, J.C., Canales-Vázquez, J., Marrero-López, D., Savvin,

S.N., Nunez, P., Dos Santos-García, A.J., Sánchez-Bautista, C., Peña-
Martínez, J. Carbon. 2010, 48, 3964-3967.
25. Jacobs, .CB., Peairs, M.J., Venton, B.J. Anal. Chim. Acta. 2010, 662,
105-127.
178
26. Mamuru, S.A., Ozoemena, K.I., Fukuda, T., Kobayashi, N., Nyokong, T.
27. Chen, Y., Lin, Y., Liu, Y., Doyle, J., He, N., Zhuang, X.D., Bai, J.R., Blau
W.J. J. Nanosci. Nanotech. 2007, 7, 1268-1283.
28. Bandaru, P.R. J. Nanosci. Nanotech. 2007, 7, 1239-1267.
29. Inagaki, M., Kaneko, K., Nishizawa, T. Carbon. 2004, 42, 1401-1417.
30. Zagal, J.H., Griveau, S., Ozoemena, K., Nyokong, T. Bedioui, F. J.

Nanosci Nanotech. 2009, 9, 2201-2214.
31. Ozoemena, K.I., Nyokong, T., Knosi, D., Chambrier, I., Coock, M.J.
32. Apetrei, C., Nieto, M., Rodríguez-Méndez, M.L., de Saja, J.A. J.

33. Khene, S. Nyokong, T. Electroanalysis. 2011, 23, 1901-1911.
34. Rodríguez-Méndez, M.L., Gay, M., Apetrei, C., de Saja, J.A.

35. Sugimoto, H., Higashi, T., Mori, M. J. Chem. Soc., Chem. Commun.
1983, 11, 622-623.
179
CHAPTER 8
Electronic Tongue Formed by Sensors and
Biosensors Containing Phthalocyanines as
Electron Mediators. Application to the Analysis of
Red Grapes.
Based on the good performance of carbon paste electrodes containing

phthalocyanines and their excellent electron mediator behavior, the aim
of this chapter was to develop a hybrid multisensor system combining
voltammetric sensors and biosensors and to evaluate its capability to
discriminate musts prepared from different varieties of grapes according
to their polyphenol and sugar content.
As stated in the previous chapter, phthalocyanines are mixed with the

carbonaceous material and the binder, forming the matrix of the sensor.
In order to prepare biosensors, enzymes need to be included in the
system. In this case, the technique used to immobilize the enzyme was
drop casting followed by a cross-linking process.
As the goal of the work was to analyze musts, enzymes devoted to the
analysis of phenols and sugars were included in the array.
In this chapter, three different arrays of sensors have been developed

and combined: an array of sensors based on phthalocyanines (MPc-CPE)
and two arrays of biosensors based on tyrosinase and glucose oxidase:
(MPc-Tyr-CPE and MPc-GOx-CPE respectively) in which phthalocyanines
are used as electron mediators.
The sensors have been immersed in model solutions of phenols

(catechol) and sugars (glucose) and the responses have been evaluated.
Then, the array has been used to analyze musts prepared from different
varieties of grapes. The capability of discrimination of the sensor array
will be investigated using Principal Component Analysis (PCA). This work
has been published in Journal of Porphyrins and Phthalocyanines.
8.1 INTRODUCTION
During the last years efforts have been made to develop a new type of
instrument for the analysis of liquids, the electronic tongue (ET)1-4.
According to the IUPAC, an ET is a multisensor system formed by an
array of low-selective sensors combined with advanced mathematical
procedures for signal processing based on pattern recognition and/or
multivariate data analysis5. These instruments provide global
183
information about the sample instead of giving information of particular

constituents. Due to this holistic approach, ETs are particularly
interesting for the analysis of complex liquids such as wines6-9. Most of
the works in the field of electronic tongues involve signal generation
from electrochemical sensors including potentiometric, impedimetric or
voltammetric sensors9-12.
In the last years, our group has developed an ET based on voltammetric

sensors chemically modified with phthalocyanines (MPc)7,13-15. Sensors
based on phthalocyanines show interesting sensing properties due to
their rich electrochemical and electrocatalytic behavior, and because
their properties can be tuned by chemical modification16-19. In addition,
sensors can be prepared using a variety of approaches that include
simple methods such as carbon paste technique or more sophisticated
methods that produce nanostructured devices11,13,20.
The voltammetric signals obtained from electrodes modified with

phthalocyanines show redox peaks associated with the studied solution
(as in classical metallic or carbonaceous electrodes), but in the modified
electrodes, the position and intensity of such redox processes are
modulated by the interaction between the solution and the
phthalocyanine (i.e. by the electrocatalytic behavior of the
phthalocyanine)18. Phthalocyanines have also demonstrated their ability
to act as electron mediators in voltammetric biosensors. In such devices,
phthalocyanines are used to facilitate the electron transfer between the
enzyme (e.g. tyrosinase) and the electrode surface21-22.
Arrays of sensors prepared from different phthalocyanine molecules

have shown a high degree of cross-selectivity and have been successfully
used to analyze model solutions of basic tastes and other complex
liquids of industrial interest such as wines4,7,14-15.
The cross-selectivity of a multisensor system could be improved by

constructing hybrid arrays combining low-specific sensors, with
biosensors containing enzymes selective to specific compounds of
interest. This can be of particular importance in the field of viticulture
184
and enology where the content of certain compounds is directly related

to the quality of wines and also to the quality of grapes from which
wines are prepared23,24. For instance, the quality of grapes is usually
established by measuring their sugar content and acidity. In addition, it
is becoming evident that the quality of grapes is also closely associated
with other parameters such as their phenolic composition25,26.

Grapes of five different varieties were harvested in 2012 in Valladolid
region (Spain). The varieties included were Tempranillo, Garnacha,
Cabernet-Sauvignon, Prieto Picudo and Mencía. Samples were selected
and harvested by the ITACYL and Bodega Cooperativa de Cigales (Spain).
The Enological Centre of Castilla y León carried out the chemical analysis
of the grapes and prepared the corresponding musts. Grapes were

analyzed following international regulations27.The results are collected
in Table 8.1.
8.2.2- Reagents and solutions

All reagents were of high purity and used without further purification
(Sigma–Aldrich). All solutions were prepared with deionized water (18.2
M·cm-1 resistivity, Milli-Q, Millipore). A 70 g·mL−1 solution of enzymes
(tyrosinase and glucose oxidase) in buffer phosphate 0.01 mol·L-1 (pH
7.0) was used for the enzyme immobilization. 10-3 mol·L-1 solutions of
catechol and glucose were prepared in buffer phsophate (pH 7.0) and
buffer citrate (pH 3.1).
Cobalt (II), copper (II) and zinc (II) phthalocyaninates (CoPc, CuPc, ZnPc)
were purchased from Sigma Aldrich. The lutetium (III)
bisphthalocyaninate (LuPc2) was synthesized and purified in its neutral
radical state following previously published procedures28.
185
8.2.3- Preparation of the electrodes
8.2.3.1- Array of sensors formed by Carbon Paste Electrodes (CPE)

modified with phthalocyanines (MPc-CPE)
An array formed by five CPE electrodes was prepared. It was formed by
three electrodes modified with metalophthalocyanines including CoPc
(CoPc-CPE), CuPc (CuPc-CPE), ZnPc (ZnPc-CPE), one electrode modified
with LuPc2 (LuPc2-CPE) and one unmodified carbon paste electrode (C-
CPE). CPEs were prepared as previously reported by mixing graphite
powder and the corresponding phthalocyanine (15 %, w/w). Nujol was
used as binder of the composite mixture. Pastes were packed into the
body of a 1 mL plastic syringe and compressed. A metallic copper wire
was used as a contact29.
Table 8.1. Results of chemical analysis of the grapes.

Polyphenol
Index (TPI)
Total
20
15
12
15
15
Potassium
(mg·L )
-1
1770
1830
1820
1450
1340
Tartaric
(g·L )
5.23
7.62
5.03
3.88
6.87
-1
acid
(g·L )
Malic
2.03
0.83
1.53
1.68
0.53
-1
acid
186
acidity
(g·L )
Total
4.14
4.94
4.26
3.48
4.37
-1
4.18
3.59
4.17
3.71
3.42
pH
Degree
(17,5)
13.60
13.15
12.30
10.60
13.00
Degree
(16,8)
14.15
13.66
12.76
11.00
13.52
Sugar
(g·L )
238.2
229.9
214.8
185.1
227.6
-1
Density
1.0098
1.0965
1.0906
1.0791
1.0956
-1)
(g·L
24.0
23.3
22.0
19.4
23.1
Brix
Tempranillo
Variety of
Sauvignon
Garnacha
Cabernet
grape
Mencía
Picudo
Prieto
8.2.3.2- Array of biosensors formed by CPE electrodes modified with

phthalocyanines and tyrosinase (MPc-Tyr-CPE)
Five CPE sensors modified with phthalocyanines were prepared as
described in the previous paragraph. The enzyme, tyrosinase (Tyr), was
then immobilized on the surface of those electrodes using the drop
casting technique followed by cross-linking. For this purpose 10 L of
0.01 mol·L-1 phosphate buffer (pH 7.0) containing 70 g·mL−1 of enzyme,
were deposited onto the electrode surface. After drying at room
temperature for 45 minutes, the biosensors were placed in a saturated
glutaraldehyde vapor for 30 minutes and dried in air for 15 minutes at
room temperature. The biosensors were then rinsed with phosphate
buffer to remove any unbound enzyme and stored at 4oC. Using this
method, five biosensors were prepared (C-Tyr-CPE; CoPc-Tyr-CPE, CuPc-
Tyr-CPE, ZnPc-Tyr-CPE and LuPc2-Tyr-CPE)30.
187
8.2.3.3- Array of biosensors formed by CPE electrodes modified with

phthalocyanines and glucose oxidase (MPc-GOx-CPE)
Five CPE sensors modified with phthalocyanines were prepared as
described in previous paragraphs but in this case, 10 L of 0.01 mol·L-1
phosphate buffer (pH 7.0) containing 70 g·mL−1 of glucose oxidase
(GOx), was immobilized on the surface of those electrodes using the
drop casting technique followed by the cross-linking above described.
Using this method, five biosensors were prepared (C-GOx-CPE; CoPc-
GOx-CPE, CuPc-GOx-CPE, ZnPc-GOx-CPE and LuPc2-GOx-CPE).
8.2.3.4- Electrochemical measurements and data analysis

Electrochemical experiments were performed using a three electrode
system with Ag|AgCl/3M as reference electrode, 1 cm2 platinum plate as
counter electrode and a sensor or a biosensor as working electrode (the
sensors forming the array were polarized sequentially). The potentiostat
used was an EG&G PARC, Model 273 potentiostat/galvanostat
(Princeton Applied Research Corp). Voltammograms were carried out at
a scan rate of 0.1 V·s-1 and the working range was adapted to the nature
of each class of electrodes. Seven repetitions per sample were
measured. In order to reduce the high dimensionality of the recorded
signals (samplessensorspotentials), a pre-processing stage was

required. For this, a feature extraction tool based on “bell-shaped-
windowing” curves called “kernels” was used to compress the
information from the original signals and to extract meaningful data
from the readings31,32. Then, the obtained coefficients (10 coefficients
per curve) were used as the input variables for Principal Component
Analysis (PCA) used for recognition of sample patterns and dis-
(similarities) between varieties of grapes.

As stated in the introduction section, the quality of grapes is mainly
related to their sugar concentration and their polyphenolic content. For
this reason, an array of CPE sensors combined with two arrays of CPE
biosensors purposely dedicated to detect the sugar and polyphenolic
content of grapes has been developed.
188
In a first step of this chapter, the response of the three arrays (MPc-CPE
sensors, MPc-Tyr-CPE biosensors modified with tyrosinase and MPc-
GOx-CPE biosensors modified with glucose oxidase) were tested
separately using two model solutions: 10-3 mol·L-1 glucose solution to
evaluate the capability of the sensors to detect sugars and 10-3 mol·L-1
catechol solution to evaluate the capability to detect phenols. The
responses of the three arrays versus catechol and glucose were also
used to evaluate the electrocatalytic properties of the different
phthalocyanines and their capabilities as electron mediators. Once
characterized, the three arrays were exposed to musts prepared from
grapes of different varieties. Finally, the three arrays were combined
and the capability of the hybrid array formed by sensors and biosensors
was evaluated.
8.3.1- Response of the array of CPE sensors modified with

phthalocyanines (MPc-CPE)
The response of the bare C-CPE electrode towards 10-3 mol·L-1 catechol
in buffer phosphate (pH 7.0) showed the expected anodic (0.5 V) and
cathodic (0.05 V) waves associated with the oxidation of the o-
dihydroquinone to benzoquinone (E1/2= 0.45 V) (Figure 8.1). CoPc-CPE,
CuPc-CPE and ZnPc-CPE showed a similar response than C-CPE however,
the intensity of the redox peaks related to the phenol increased their
intensity (from 4 A in C-CPE to ca. 13-15 A in CoPc, CuPc and ZnPc).
Notice that in the case of the CoPc-CPE electrode, a peak due to the
oxidation of Co was also clearly observed at 0.2 V. The most important
electrocatalytic effect was observed when using the LuPc2-CPE sensor.
The anodic peak produced by the oxidation of the phenol appeared
shifted to lower potentials (observed at 0.4 V instead of 0.5 V) and the
cathodic wave was at 0.15 V (E= 0.25V). According to these results it
can be confirmed that in the LuPc2-CPE sensor the overpotential is
reduced and the reversibility is clearly improved. It is also important to
remark that the redox activity of the LuPc2 associated with LuPc2/LuPc2-
was observed at -0.2 V (anodic) and -0.4 V (cathodic).
189
Figure 8.1. Response of CPE electrodes modified with phthalocyanines to

-3 -1
10 mol·L catechol in buffer phosphate (pH 7.0). a) bare C-CPE, b) CoPc-CPE, c)
-1
CuPc-CPE, d) Zn-CPE and e) LuPc2-CPE. Scan rate, 0.1 V·s .
CPE sensors modified with phthalocyanines were also used to detect

glucose. It is well known that glucose is difficult to detect using non-
catalytic electrodes, because the oxidation of glucose to gluconic acid
requires large overpotentials33. This is the reason why the C-CPE
electrode could not detect the presence of glucose and the
voltammograms were characterized by a complete absence of peaks.
When using CPE chemically modified with ZnPc-CPE or CuPc-CPE the
response towards glucose was weak and only one small cathodic wave
was observed at ca. -0.5 V (Figure 8.2.a). In contrast, when electrodes
chemically modified with CoPc-CPE and LuPc2-CPE were tested, a drastic
increase in the cathodic response at ca. -0.8 V was observed (from -10
A in C-CPE and ZnPc-CPE, to -50, -55 A in CoPc-CPE and LuPc2-CPE)
(Figure 8.2.b).
190
-3
Figure 8.2. Response of CPE electrodes modified with phthalocyanines to 10
-1
mol·L glucose in buffer phosphate (pH 7.0) a) ZnPc-CPE and b) CoPc-CPE. Scan
-1
rate, 0.1 V·s .
According to these results, it can be concluded that the chemical

structure of the phthalocyanine plays a critical role in detection of
catechol and glucose. It is particularly remarkable that CoPc and LuPc2
(which have redox reactivity in the working range) show enhanced
electrocatalytic properties facilitating the electrochemical
oxidation/reduction of both catechol and glucose.
Taking into account that grapes and musts obtained from the berries
have a pH close to 3.5, the response of the CPE sensors modified with
phthalocyanines towards catechol and glucose was also tested at an
acidic pH obtained from a buffer citrate (pH 3.1). The effect of the acidic
pH is illustrated in Figure 8.3 where the responses of the LuPc2-CPE
towards catechol and glucose at pH 7.0 and at pH 3.1 are shown.
It is well known that pH affects the electrochemical response of phenols

and when decreasing the pH, the oxidation of phenols occurs at higher
potentials34,35. This effect was also observed in our phthalocyanine-
based electrodes (Figure 8.3.a). Moreover, at pH 3.1 the intensity of the
voltammograms increased drastically. This increase in the sensitivity
could be due to the protonation of the phthalocyanine molecule and can
be a good advantage in the detection of phenols in musts.
191
The effect of the pH in the detection of glucose was the opposite and
the intensity of the peaks decreased at pH 3.1 (Figure 8.3.b). This result
is in accordance with previously published results that conclude that
gluconic acid in its free form reversibly inhibits the oxidation process in
acidic media36. Nevertheless, the inhibition is only partial and the
intensity of the signals is good enough for applications in the field of
wines and musts.
-3
Figure 8.3. Cyclic voltammograms of LuPc2-CPE electrodes immersed in a) 10
-1 -3 -1
mol·L catechol and b) 10 mol·L glucose. Blue lines correspond to
measurements in buffer phosphate (pH 7.0) and red lines correspond to
-1
measurements registered in buffer citrate (pH 3.1). Scan rate, 0.1 V·s .
8.3.2- Response of the array of biosensors formed by CPEs

modified with phthalocyanines and tyrosinase (MPc-Tyr-CPE)
In order to improve the selectivity of the electronic tongue towards
polyphenols, an array formed by biosensors containing tyrosinase was
prepared. Tyrosinase is involved in the oxidation of diphenols to the
corresponding quinoid form. Then, the formed benzoquinone can be
reduced electrochemically at the electrode surface. Phthalocyanines
used as electron mediators can facilitate the transfer of electrons from
the enzyme to the electrode, improving the performance of the
biosensors21,22,37.
The response of the array MPc-Tyr-CPE electrodes towards catechol is

illustrated in Figure 8.4. The influence of the pH in the response is also
illustrated in the figure.
192
In all MPc-Tyr-CPE electrodes, voltamograms registered at pH 7.0,

showed the electrochemical reduction of the previously formed
benzoquinone in the -0.4 V region (Figure 8.4.a). The reduction of the
quinone to the phenolic form, shifted to lower potentials in the
presence of phthalocyanines (results not shown). It is particularly
remarkable the shift observed in CoPc-Tyr-CPE and LuPc2-Tyr-CPE where
this process occured at ca. 0.0 V. The electrochemical behavior of these
two electrodes was also different from the rest, because redox
processes of the phthalocyanines which occur in the working range
(CoI/CoII and LuPc2-/LuPc2) can facilitate the electron transfer to the
electrode.
The pH of a solution has several effects on the structure and activity of

enzymes. The optimal pH for tyrosinase is about 6-737. The acidic pH of
grapes and musts (close to 3.5) could inhibit the enzymatic activity. In
Figure 8.4.a and 8.4.b the voltammograms of the C-Tyr-CPE biosensor
towards catechol dissolved in buffer phosphate (pH 7.0) and in buffer
citrate (pH 3.1) are compared. The enzymatic reduction of the
benzoquinone shifted from -0.35 V at pH 7 to +0.17 V at pH 3.1 while the
intensity of the signal decreased. The most remarkable difference is that
at pH 3.1, a new peak was observed at ca. 0.8 V. The acidity dissociation
constants (pKa) for catechol are both above pH 7, but the pKa for the
catechol radical is close to 438. Thus, the pH can affect the dissociation
state and hence the electroxidation/reduction mechanism of the
catechol, facilitating the electrochemical oxidation at 0.8 V.
As shown in Figure 8.4.c-f, similar trends were observed in all the MPc-
Tyr-CPE sensors immersed in catechol solved in citric acid (pH 3.1).
193
Figure 8.4. Response of CPE electrodes modified with phthalocyanines and

-3 -1
tyrosinase towards 10 mol·L catechol. a) C-Tyr-CPE in buffer phosphate (pH
7.0), b) C-Tyr-CPE in buffer citrate (pH 3.1), c) CoPc-Tyr-CPE in buffer citrate (pH
3.1), d) CuPc-Tyr-CPE in buffer citrate (pH 3.1), e) ZnPc-Tyr-CPE in buffer citrate
-1
(pH 3.1) and f) LuPc2-Tyr-CPE in buffer citrate (pH 3.1). Scan rate, 0.1 V·s .
8.3.3- Response of the array of biosensors formed by CPEs

modified with phthalocyanines and glucose oxidase (MPc-GOx-
CPE)
With the aim of increasing the selectivity towards glucose, a third array
formed by biosensors containing glucose oxidase was prepared. In
electrochemical biosensors based on the glucose oxidase, the enzymatic
oxidation of glucose produces gluconic acid and hydrogen peroxide with
consumption of oxygen39. The reduced glucose oxidase is not able to
transfer electrons to conventional electrodes because the distance
between its redox centers and the electrode surface exceeds the
distance across which electrons are transferred at sufficient rates.
Therefore, electrical communication between the redox centers of this
enzyme and electrodes require the presence of electron mediators. A
suitable mediator for the GOx enzyme must have a more positive redox
potential than the redox potential of the coenzyme FAD (ca. −0.4 V
versus Ag/AgCl)40. It has been reported that some phthalocyanines such
as CoPc can act as electron mediators of the GOx enzyme41. In this work,
194
the electron mediator properties of CoPc, CuPc, ZnPc and LuPc2 have
been analyzed and compared.
At pH 7.0 (buffer phosphate), the presence of phthalocyanines increased

the intensity of the response towards glucose with respect to the signal
obtained using an electron mediator free electrode. This is illustrated in
Figure 8.5.a where the responses of the phthalocyanine-free C-GOx-CPE
electrode and CuPc-GOx-CPE and ZnPc-GOx-CPE electrodes are
compared. However, as shown in Figure 8.5.b, the difference between
the response of an electrode modified with both phthalocyanine and
GOx and one electrode modified only with a phthalocyanine is not so
dramatic.
Figure 8.5. a) Voltammetric response of CPE electrodes modified with

-3 -1
phthalocyanines and glucose oxidase towards glucose 10 mol·L in buffer
phosphate (pH 7.0), C-GOx-CPE (blue line), CuPc-GOx-CPE (red line) and ZnPc-
GOx-CPE (green line). b) Cyclic voltammograms of C-GOx-CPE (blue line), LuPc2-
-3 -1
CPE (red line) and LuPc2-GOx-CPE (green line) immersed in glucose 10 mol·L
-1
in buffer phosphate (pH 7.0). Scan rate, 0.1 V·s .
On the other hand, the optimal pH for glucose oxidase is 5.5 with broad
range 4 – 7. Surprisingly, the voltammograms registered in buffer citrate
at pH 3.1 (closest to the pH found in wines and musts) showed a clear
increase in the intensity of the responses, that was particularly
noticeable for the LuPc2-GOx-CPE (Figure 8.6). This enhancement can be
the result of the protonation of the phthalocyanine ring that modifies
195
the electrochemical properties of the phthalocyanines13,29 improving

their electron mediator activity.
Figure 8.6. Response of LuPc2-CPE(blue line) and LuPc2-GOx-CPE (red line) to

-3 -1 -1
10 mol·L glucose in buffer citrate (pH 3.1). Scan rate, 0.1 V·s .
8.3.4- Response towards musts

The sensors and biosensors described in the previous sections were
exposed to musts prepared from grapes of different varieties. In order to
decrease the complexity of the samples, musts were diluted 50% in
water (Figure 8.7). Voltammograms were dominated by the redox
response of the phenolic groups present in musts that appeared as
anodic peaks in the 0.4-0.8 V region and the corresponding cathodic
waves in the 0.35 V region. Obviously, in such a complex media, the
peaks were broader than in the corresponding model solutions and a
variety of other small and not well-defined peaks were observed. In all
cases, a cathodic peak in the -0.5 V region was also observed that could
be associated with the glucose but also with the presence of protons.
The intensity and position of the peaks depended on the nature of the
phthalocyanine used as modifier. As expected, in the case of sensors
modified with CoPc and LuPc2 the peaks increased their intensity and
shifted to lower potentials.
The presence of tyrosinase or glucose oxidase also introduced important

modifications in the electrochemical responses. The intensities and
196
positions of the peaks were related to the total polyphenols index and
the content of glucose measured by chemical methods that in turn
depend on the grape variety (Table 8.1).
In summary, each electrode provided a different response towards the

same must sample and an important degree of cross-selectivity was
attained.
Figure 8.7. Response of CPE electrodes modified with phthalocyanines towards

Garnacha must diluted 50% in water. a) CoPc-CPE, b) LuPc2-CPE, c) CoPc-Tyr-
CPE, d) LuPc2-Tyr-CPE, e) CoPc-GOx-CPE and f) LuPc2-GOx-CPE. Scan rate, 0.1
-1
V·s .
197
8.3.5- Data treatment

As demonstrated in the above sections, the CPE sensors and biosensors
showed complex voltammograms which contain information about the
musts. For this reason, voltammograms can be used to discriminate
must samples using Principal Component Analysis (PCA)15,29,43.
In order to evaluate the capability of the three arrays to discriminate the

variety of grapes, they were immersed in each must and 7 replicas of
each sample were registered. Then, voltammograms were pre-treated
to decrease the number of variables used in the PCA calculations by
means of the kernel method. The 10 values obtained from each
voltammogram were scaled between the maximum and minimum
values to discard range current effects and standardized (i.e. mean
value=0, standard deviation=1) to build the matrix used for the pattern
recognition techniques.
The three arrays were able to discriminate the five musts. This is
illustrated in Figure 8.8 where PCA scores plot and loadings plot
corresponding to the musts analyzed with the array of MPc-Tyr-CPE is
shown. As observed in the scores plot (Figure 8.8.a), the percentage of
variance explained using two principal components was 97.28%, that is,
almost the total information of the model was contained in two principal
components. The array discriminated the musts according to their
polyphenol content. For this reason Tempranillo must which has the
highest TPI (see Table 8.1) appears on the right side of the graph, in the
positive PC1, far apart from the rest of the musts, whereas, Mencia must
that presents the lowest TPI content, appears at negative PC1 on the left
side of the diagram. Prieto Picudo, Garnacha and Cabernet-Sauvignon
musts, appeared close one to each other because they have a similar
polyphenolic content (TPI=15). As observed in Figure 8.8.b where the
loading plot is presented, the loadings provided by each sensor (ten
kernels per sensor) appear in different regions of the loading plots
confirming that the central metal ion coordinated to the phthalocyanine
ring plays an important role in the selectivity of the sensors. This result
demonstrates that sensors based on different phthalocyanines provide
198
complementary information and show an important degree of cross

selectivity. Similar results were obtained when using the array of MPc-
GOx-CPE biosensors where the array discriminates according to the
sugar content.
Figure 8.8. a) PCA scores plot corresponding to the classification of the musts.
Analysis performed using the CPE modified with phthalocyanines and
tyrosinase. 1-Tempranillo, 2-Prieto Picudo, 3-Mencía, 4-Cabernet-Sauvignon
and 5-Garnacha. b) Loading plot of the PCA (ten variables were extracted per
sensor using the kernel method).
Finally, the three arrays of MPc-CPE, MPc-Tyr-CPE and MPc-GOx-CPE

were combined in a single multisensory system and the PCA was
performed. The results are shown in Figure 8.9 where the PCA scores
plot and loading plot are presented. In this case, due to the increase in
the number of sensors (and variables) the three principal components
bring a noticeable amount of information. The first principal component
(PC1) explains the 54.16% of the information, the second principal
component (PC2) the 25.92% and the third one (PC3) the 9.46%.
As observed in the figure, when using the complete set of sensors a

better discrimination is achieved and all the musts appear well
separated. The combined array containing non specific MPc-CPE sensors
with biosensors sensitive to polyphenols (MPc-Tyr-CPE) and glucose
(MPc-GOx-CPE) provided a specific signature for each must that allowed
discriminating the samples according to the variety of grape. It is also
199
remarkable that sensors forming the array had an excellent

complementarity as observed in the loading plot (Figure 8.9.b).
Although the system provides global information about the sample, it is

important to notice that some correlations with the composition can be
found in the PCA. For instance, the must prepared from grapes of
Tempranillo variety appears on the left and up side of the graph, well
apart from the rest of the musts. This result is in good accordance with
the results obtained by means of chemical analysis shown in Table 8.1,
where it can be observed that the grapes of the variety Tempranillo
have a TPI and a glucose content (and a Brix degree) clearly higher than
the levels observed in the other musts. Musts prepared from Mencia,
Prieto Picudo and Cabernet appear in the bottom part of the graph,
indicating that their TPI and Brix degree is lower. As the quality of the
grapes (and of the corresponding musts) depends on the TPI and the
sugar content, this result indicates that our bioelectronic tongue could
be an interesting instrument to evaluate the quality of berries.
Figure 8.9. a) PCA scores plot and b) loading plots (ten variables were extracted
per sensor using the kernel method) corresponding to the classification of the
musts. Analysis performed using the array formed by the CPE modified with
phthalocyanines, CPE modified with phthalocyanines and tyrosinase and CPE
modified with phthalocyanines and glucose. 1-Tempranillo, 2-Prieto Picudo, 3-
Mencía, 4-Cabernet-Sauvignon and 5-Garnacha.
200
8.4 CONCLUSIONS
An electronic tongue based on voltammetric sensors and biosensors
containing phthalocyanines has been developed and used to
discriminate musts prepared from different varieties of grapes.
The important degree of cross-selectivity of the MPc-CPE array towards

catechol and glucose was related to the electrocatalytic properties of
the phthalocyanines used. It was demonstrated that the electrocatalytic
effect of CoPc and LuPc2 towards catechol and glucose was superior to
the observed in ZnPc and CuPc. In addition, the performance of the
arrays of biosensors modified with tyrosinase and glucose oxidase (MPc-
Tyr-CPE and MPc-GOx-CPE) was also improved in the presence of
phthalocyanines as electron mediators.
The electrochemical responses towards musts were associated with the

content of polyphenols and glucose of the samples. A multivariate data
treatment was carried out to explore the capability of discrimination of
the array. PCA showed a good discrimination according to the
composition of the musts in terms of polyphenolic content and sugar
concentration. The capability of discrimination of the system was
improved by combining the MPc-CPE sensors, the MPc-Tyr-CPE and
MPc-GOx-CPE biosensors to form an array of sensors with high cross-
selectivity. These results confirm the possibility of using this new array
of sensors to study the quality of grapes.
In summary, this multisensor system containing sensors and biosensors

could give the classical global information about the characteristics of
the samples plus complementary information about sugars or
polyphenolic content that could help to decide important issues such as
the retail price (marked by the quality) or the appropriate harvesting
date.
201
References
1. Winquist, F. Microchim. Acta. 2008, 163, 3-10.
2. Riul, A.Jr., Dantas, C.A.R., Miyazaki, C.M., Oliveira, O.N.Jr. Analyst.

2010, 135, 2481-2495.
3. Del Valle, M. Electroanal. 2010, 22, 1539–1555.
4. Rodríguez-Méndez, M.L., Parra, V., Apetrei, C., Villanueva, S., Gay, M.,
Prieto, N., Martínez, J., de Saja, J.A. Microchim. Acta. 2008, 163, 23-31.
5. Vlasov, Y., Legin, A., Rudnitskaya, A., Di Natale, C. D'Amico, A. Pure

Appl. Chem. 2005, 77, 1965-1983.
6. Cetó, X., Gutiérrez, J.M., Moreno-Barón, L., Alegret, S. del Valle, M.

Electroanal. 2011, 23, 72-78.
7. Gay, M., Apetrei, C., Nevares, I., del Alamo, M., Zurro, J., Prieto, N., de
Saja, J.A., Rodriguez-Mendez, M.L. Electrochim. Acta. 2010, 55, 6782-
6788.
8. Rudnitskaya, A., Rocha, S.M., Legin, A., Pereira, V., Marques, J.C. Anal.
Chim. Acta. 2010, 662, 82-89.
9. Moreno i Codinachs, L., Kloock, J.P., Schöning, M.J., Baldi, A., Ipatov,
A., Bratov, A. Jimenez-Jorquera, C. Analyst. 2008, 133, 1440-1448.
10. Ciosek, P., Wróblewski, W. Sensors. 2011, 11, 4688-4701.
11. Furini, L.N., Feitosa, E., Alessio, P., Shimabukuro, M.H., Riul, A.Jr.,
Constantino, C.J.L. Mater. Sci. Eng., C. 2013, 33, 2937-2946.
12.Winquist, F., Olsson, J., Eriksson, M. Anal. Chim. Acta. 2011, 683, 192-
197.
13. Apetrei, C., Nieto, M., Rodriguez-Mendez, M.L., de Saja, J.A. J.

202
14. Apetrei, I.M., Rodriguez-Mendez, M.L., Apetrei, C., Nevares, I., del
Alamo, M., de Saja, J.A. Food Res. Int. 2012, 45, 244-249.
15. Apetrei, C., Apetrei, I., Nevares, I., del Alamo, M., Parra, V.,
2594.
16. Zagal, J.H., Griveau, S., Silva, J.F., Nyokong, T., Bediou, F. Coordin.
Chem. Rev. 2010, 254, 2755-2791.
17. Gay, M., Rodriguez-Mendez, M.L., de Saja, J.A. Langmuir. 2010, 26,
19217-19224.
18. Rodriguez-Mendez, M.L., Gay, M. de Saja, J.A. J. Porphyrins

19. Zagal, J.H., Griveau S, Santander-Nelli M and Bedioui F. J. Porphyrins

20. Selektor, S.L., Shokurov, A.V., Raitman, O.A., Sheinina, L.S., Arslanov,
V.V., Birin, K.P., Gorbunova, Y.G., Tsivadze, A.Y. Colloid J. 2012, 74, 334-
345.
21. Apetrei, C., Alessio, P., Constantino, C.J.L., de Saja, J.A., Rodriguez-
Mendez, M.L., Pavinatto, F.J., Fernandes, E.G., Zucolotto, V., Oliveira,
O.N.Jr. Biosens. Bioelectron. 2011, 26, 2513-2519.
22. Pavinatto, F.J., Fernandes, E.G.R., Alesio, P., Constantino, C.J.L., de

Saja, J.A., Zucolotto, V., Apetrei, C., Oliveira, O.N.Jr., Rodriguez-Mendez,
M.L. J. Mater. Chem. 2011, 21, 4995-5003.
23. Riberéau-Gayon, P., Glories, Y., Maujean, A., Dubordieu, D.

Compuestos fenólicos. Tratado de enología. Química del vino,
estabilización y tratamientos. Vol 2, 1999. Ed. Mundi-prensa. Madrid
(Spain)
24. Revilla, E., Alonso, E., Kovac, V. Wine: Nutritional and Therapeutic
Benefits. The content of catechin and procyanidins in grapes and wines
203
as affected by agroecological factors and technological practices. 1997,

Ed. American Chemical Society. Washington DC (USA)
25. Boulton, R. Am. J. Enol. Vitic. 2001, 52, 67-87.
26. Glories, Y. Caracterisatión du potentiel phenolique : adaptation de la

vinification. Progrès Agricole et Viticole. Vol.118. 2001, 347-350.
27. Compendium of International Methods of Analysis of Wines and

Musts. Vol. 2. OIV, 2013. Paris (France).
28. Linaje, M., Quintanilla, M.C., Gonzalez, A., del Valle, J.L., Alcaide, G.
Rodriguez-Mendez, M.L. Analyst. 2000, 125, 341-346.
29. Rodriguez-Mendez, M.L., Gay, M., Apetrei, C., de Saja, J.A.

30. Li, Y.G., Zhou, Y.X., Feng, J.L., Jiang, Z.H., Ma, L.R. Anal.Chim.Acta.
1999, 382, 277-282.
31. Cetó, X., Céspedes, F., del Valle, M. Microchim. Acta. 2013, 180, 319-
330.
32. Parra, V., Hernando, T., Rodriguez-Mendez, M.L., de Saja, J.A.

33. Park, S., Boo, H., Chung, T.D. Anal. Chim. Acta. 2006, 556, 46-57.
34. Martinez, S., Valek, L., Petrovic, Z., Metikos-Hukovic, M., Piljac, J. J.
Electroanal. Chem. 2005, 548, 92-99.
35. Kilmartin, P.A., Zou, H., Waterhouse, A.L. Am. J. Enology and
Viticulture. 2002, 53, 294-302.
36. Abbadi, A., Makkee, M., Visscher, W., van Veen, J.A.R., van Bekkum
H. J. Carbohyd. Chem. 1993, 12, 573-587.
37. Yin, H.S., Zhou, Y.L., Ai, S.Y. J. Electroanal. Chem. 2009, 626, 80-88.
204
38. Chang, T.S. Int. J. of Mol. Sci. 2009, 10, 2440-2475.
39. Scheweigert, N., Zehnder, A.J.B., Rik, I.L. Environ. Microbiol. 2001, 3,
81-91.
40. Degani, Y., Heller, A. J. Phys. Chem. 1987, 91, 1285-1289.
41. Harper, A., Anderson, M.R. Sensors. 2010, 10, 8248-8274.
42. Wang K, Xu JJ and Chen HY, Biosens. Bioelectron. 2005, 20, 1388-
1396.
43. Apetri, C., Rodriguez-Mendez, M.L., de Saja, J.A. Electrochim.Acta.

2008, 53, 5867-5872.
205
CHAPTER 9
The Advantages of Disposable Screen-Printed
Biosensors in a Bioelectronic Tongue for the
Analysis of Grapes
Despite the good results obtained with the previous sensors either
nanostructured or CPE, the use of liquid electrochemical cells and heavy
potentiostast restrict their use to the laboratory. It would be desirable to
develop a portable multisensor system using disposable electrodes that
could be used in the vineyard.
Screen Printed Electrodes (SPE) are a good option for this kind of system
as they are low cost, disposable and portable. They include the
reference electrode, the counter electrode and the working electrodes
according to the design needed on the same device. The working
electrode can be made of a wide variety of materials including glassy
carbon, platinum, gold and nanoparticles. In addition, they can be easily
modify by means of casting technique and allow user to work with a
small amount of sample.
The aim of this chapter was to develop a bioelectronic tongue, based on

an array of disposable SPEs dedicated to the analysis of grapes. For this
purpose, six SPE sensors made with different materials (M-SPE) have
been selected, including three classical electrode materials: graphite (C-
SPE), platinum (Pt-SPE) and gold (Au-SPE), two nanostructured
electrocatalytic materials: graphene (GPH-SPE) and nickel oxide
nanoparticles (NiONP-SPE) and material with well-known electron
mediator properties: Prussian Blue (PB-SPE). Then, glucose oxidase
(GOx) or Tyrosinase (Tyr) have been immobilized on the surface of the
M-SPE electrodes to obtain M-GOx-SPE and M-Tyr-SPE biosensors that
could provide information about the glucose and phenol content
respectively. The discriminatory capability of the sensor array will be
investigated using Principal Component Analysis (PCA). This research has
been published in LWT-Food Science and Technology.
9.1 INTRODUCTION
The use of accurate techniques for the analysis of grapes is an important
need in the wine sector. As stated in previous chapters,
glucose and polyphenols are among the analytes that need to
be monitored and measured in order to guarantee the maturity of the
grapes and the quality of the final product. In the laboratory, the
209
assessment of grape maturity is performed by analyzing representative

samples by standard analytical techniques such as spectroscopy,
chromatography (to measure phenols) or refractometry (to measure the
sugar content in Brix )1. In the last few years, a variety of sensors
dedicated to the grape analysis have been developed. Most of them are
fiber-optic fluorescence sensors that measure the levels of anthocyanins
and chlorophyll2, or near-infrared optical sensors that can assess the
sugar content (oBrix)3. Electrochemical glucose and polyphenol
biosensors are a powerful alternative for monitoring such compounds
due to their specificity, high sensitivity and short response time4-6.
Typical enzymatic electrodes, based on glucose oxidase (or tyrosinase),

undergo several electrochemical steps which produce a measurable
current that is linearly related to the glucose (or polyphenol) content. In
order to improve the electron transfer, electron mediators must be
included in the sensor4,6-8. Nanotechnology has opened up new
opportunities to obtain more efficient electron mediators, such as metal
nanoparticles or nanocarbons (e.g. carbon nanotubes or graphene)5,9.
The analysis of complex mixtures such as wines or musts will be

improved by using the so-called electronic tongues (ET), a new class of
instruments that provide global information about the sample, instead
of information about specific compounds. There have been a range of
portable electronic tongues which have been produced based on a
numerous core sensing technologies including mass, optic, electrical or
electrochemical transduction10-16. Arrays of electrochemical sensors
have been successfully used to discriminate between wines with
different organoleptic characteristics17-25 or grapes with different
qualities26,27. It has also been established that arrays combining sensors
and biosensors can be advantageous, as they bring together both the
benefits of classical arrays of electrochemical sensors (which provide
global information about the sample) with the specificity of the enzyme-
substrate reaction typical of biosensors27-29
In spite of the good results obtained with electronic and bioelectronic

tongues, the use of liquid electrochemical cells and heavy potentiostats,
210
the high cost of the sensors and the need for periodic calibration restrict
their use to the laboratory. It would be desirable to develop multisensor
systems using disposable electrodes that could be used in the vineyard
block.
During recent years, the screen-printing (thick film) technology, applied

to sensor and biosensor construction, has been considerably improved.
Screen-printed electrodes (SPEs) are in fact simple to prepare, rapid and
versatile, and this technology also appears to be the most economical.
Their low cost makes possible to use them as disposable electrodes. SPE
containing a serigraphied pseudo-reference Ag/AgCl electrode, a
counter electrode and a working electrode in the same device, can be
prepared using a variety of modifiers that include metals, metal oxides,
nanocarbons, nanoparticles, phthalocyanines, conducting polymers, or
prussian blue among many others30. Moreover, enzymes can be
immobilized on modified screen-printed electrodes to take advantage of
the electron mediator properties of many of these materials for
biosensing.
The aim of this chapter is to develop a bioelectronic tongue, based on an

array of disposable screen-printed electrodes, dedicated to the analysis
of grapes. For this purpose, six SPE sensors, modified with different
materials (M-SPE) will be selected including three classical electrode
materials: graphite (C-SPE), platinum (Pt-SPE) and gold (Au-SPE), two
nanostructured electrocatalytic materials: graphene (GPH-SPE) and
nickel oxide nanoparticles (NiONP-SPE) and materials with well-known
electron mediator properties: Prussian Blue (PB-SPE). Then, glucose
oxidase (GOx) or Tyrosinase (Tyr) will be immobilized on the surface of
9.2.1- Chemicals
All chemicals and solvents were purchased from Sigma-Aldrich. Solvents
were of reagent grade and used as supplied. Deionized water was
obtained from a Millipore purifier, with a resistivity of 18.2 M·cm.
211
Glucose oxidase (GOx), from Aspergillus Niger, type VII (activity plus
0.001 kat·mg-1), and Tyrosinase (Tyr) from Mushroom (activity of 610-5
kat·mg-1), were purchased from Sigma Chemical Co. (USA).

Samples of five varieties of red grapes (Tempranillo, Garnacha,
Cabernet-Sauvignon, Prieto Picudo and Mencía) were harvested in
September 2013. For each variety, berries were collected in the
vineyards of ‘Bodega Cooperativa de Cigales’ and of ‘Instituto
Tecnologico Agrario de Castilla y León’, both located in Valladolid area
(Spain). To obtain the musts, 200 berries were crushed manually. For
each variety of grapes, this process was carried out by septuplicate,
giving a total of 35 samples. The musts were analyzed after separation
from the seeds and the peel. The Oenological Center of Castilla y León
carried out the chemical analysis of the grapes following international
regulations31. Parameters analyzed included the chemical indicators of
the sugar concentration (oBrix) and of the polyphenolic content (Total
Polyphenol Index: TPI). Total acidity (expressed as g·L-1 of tartaric acid)
and pH were also analyzed, due to the influence of the pH and the ionic
strength in the enzymatic activity. The results are collected in Table 9.1.
212
Table 9.1. Results of the chemical analysis of grapes carried out by traditional
chemical methods (seven replicas).
o Total Acidity Total Polyphenol

Grape Variety Brix pH
(g·L-1) Index
Prieto Picudo 23±2 3.61±0.08 4.8±0.2 15±3
Mencia 21±2 4.18±0.08 4.3±0.2 13±3
Tempranillo 24±2 4.20±0.08 4.1±0.2 21±4
Cabernet-Sauvignon 23±2 3.65±0.08 4.1±0.2 16±3
Garnacha 23±2 3.39±0.08 4.4±0.2 15±3
9.2.3- Screen-printed electrodes and biosensors

Six screen-printed electrodes modified with different materials (M-SPE),
purchased from Dropsens (www.dropsens.com) were used as substrates
for the deposition of enzymes. Each sensor device contained a
serigraphied pseudo-reference Ag/AgCl electrode (RE), a counter
electrode (C or Pt) (CE) and a working electrode. As observed in Table
9.2, the modifiers covering the working electrode of the M-SPE devices
included carbon (C), platinum (Pt), gold (Au), graphene (GPH), nickel
oxide nanoparticles (NiONP) and Prussian blue (PB).
The six screen-printed electrodes were modified with GOx or Tyr using
the drop casting technique followed by cross-linking. For this purpose,
10 l of a 70 g·L-1 solution of the corresponding enzyme in 0.01 mol·L-1
phosphate buffer (pH 7.0) were placed onto the working electrode
surface. After drying at room temperature for 30 minutes, cross linking
was carried out by exposing the modified working electrode to
glutaraldehyde vapors (20 g·L-1 water solution) for 20 minutes.
Throughout the process, CE and RE were protected with a mask to avoid
contamination by the enzyme or the glutaraldehyde. Using this method,
six GOx-based (M-GOx-SPE) and six Tyr-based (M-Tyr-SPE) biosensors
were prepared (Table 9.2).
213
Dropsens Sensors containing Glucose Sensors containing Tyrosinase
Modifier/Acronym
Reference Oxidase Modifier/Acronym Modifier/Acronym
DRP-110 Carbon/C-SPE Carbon/C-GOx-SPE Carbon/C-Tyr-SPE
DRP-550 Platinum/Pt-SPE Platinum/Pt-GOx-SPE Platinum/Pt-Tyr-SPE

DRP-250AT Gold/Au-SPE Gold/Au -GOx-SPE Gold/Au -Tyr-SPE
214
DRP-110GPH Graphene/GPH-SPE Graphene/GPH-GOx-SPE Graphene/GPH -Tyr-SPE
DRP-110NI NiO Nanoparticles/NiONP-SPE NiO Nanoparticles/NiO-GOx-SPE NiO Nanoparticles/NiO-Tyr-SPE
DRP-710 Prussian Blue/PB-SPE Prussian Blue/PB -GOx-SPE Prussian Blue/PB -Tyr-SPE

Table 9.2. List of the screen printed electrodes and acronyms used in this work.

Electrochemical measurements were carried out in a portable (weight
under 1 g) potentiostat µSTAT 400 (from Dropsens). Cyclic
voltammograms were registered from –0.6 V to +1.0 V at a sweep rate
of 0.05 V·s-1. The applied potential was measured versus the screen-
printed internal silver pseudo-reference electrode. Sensors were
immersed in the solution and polarized sequentially. The
electrochemical experiments were performed at a controlled
temperature of 21oC. After each measurement, sensors were discarded
and replaced by a new sensor and musts were measured randomly.
9.2.5- Statistical analysis

Curves were pre-processed through the adaptation to our case of a data
reduction technique based on predefined response “bell-shaped-
windowing” curves called “kernels”23. It is based on a compression
method32. The voltammogram curve is multiplied by 10 smooth bell-
shaped windowing functions defined as (Equation 9.1)
1
𝐾𝑖 (𝑉𝑗 ) = 𝑉𝑗−𝑐𝑖 (Equation 9.1)
1+( )2𝑏𝑖
𝑎𝑖
Where ai, bi and ci define the width, the shape and the center of the
different windowing functions Ki. Subsequently, data were integrated
with respect to voltage. After compression, each voltammogram has
been reduced to a vector of 10 variables.
Using this method, ten variables per curve were obtained which were
used as the input data source for statistical analysis. In previous works, it
has been demonstrated that this method provides similar results to
other pre-treatment processes such as the wavelet transformation
(WT)18 or genetic algorithms33.
A non-supervised multivariate method, the Principal Component

Analysis (PCA), was used to analyze the voltammetric curves and to
evaluate the discrimination capability of the system. Partial Least
215
Squares (PLS-1) regression models were built to predict the phenolic and
the sugar content of the musts.
All computations and chemometric analysis were carried out using the
software Matlab v5.3. (The Mathworks Inc., Natick, MA, USA) and
Unscrambler (Camo).
9.3.1- Measurements in glucose and catechol

As a preliminary experiment, the electrochemical response of screen-
printed sensors immersed in 0.01 mol·L-1 phosphate buffer was
analyzed. As expected, C-SPE, GPH-SPE and NiONP-SPE sensors did not
show any redox peak within the studied range. Pt-SPE and Au-SPE
showed the usual peaks related to the oxidation/reduction of water and
those associated with the adsorption of oxidized species to the surface.
Finally, PB-SPE showed the redox peaks of the hexacyanoferrate ion34.
The effect of the electrode modification with GOx on the oxidation of

glucose at the SPE surface and the role of the six materials used as
electron mediators was evaluated by performing voltammetric
experiments. Figure 9.1 compares the cyclic voltammograms of the M-
SPE and M-GOx-SPE electrodes in the presence of 0.01 mol·L-1 glucose in
phosphate buffer.
The responses of C-SPE, Pt-SPE and Au-SPE were similar to those of the
electrodes modified with the enzyme (C-GOx-SPE, Pt-GOx-SPE and Au-
GOx-SPE). The only noticeable differences were small increases in the
background currents in M-GOx-SPE and a slight increase in the current
intensity at potentials related to the electrochemical reactivity of H2O2
(at -0.4 V and lower). The curves obtained with GPH-GOx-SPE, NiONP-
GOx-SPE and PB-GOx-SPE showed a clear increase in the current
intensities of the whole curves, demonstrating that graphene, nickel
oxide nanoparticles and Prussian blue were excellent electron
mediators. It is important to remark that Prussian blue was particularly
effective in enhancing the sensitivity towards glucose. The good electro-
216
catalytic activity of such modifiers was in good agreement with results

described in the literature using other types of electrodes9,34,35.
-1 -1
Figure 9.1. Voltammetric response towards 0.01 mol·L glucose in 0.01 mol·L
phosphate buffer (pH 7.0) of screen printed electrodes modified with a)
graphene, b) nickel oxide nanoparticles and c) Prussian blue. Red lines
correspond to M-CPEs response and blue lines correspond to M-GOx-CPE
-1
curves. Scan rate, 0.05 V·s .
The electrochemical response of the M-SPE and M-Tyr-SPE sensors

towards phenols was evaluated by registering the cyclic voltammograms
in a 0.01 mol·L-1 catechol solution (in 0.01 mol·L-1 phosphate buffer pH
7.0). In all M-SPE sensors, the oxidation of catechol to the quinoid form
was observed at ca. 0.8 V. This peak was accompanied by a peak at ca.
-0.2 V produced by the reverse reaction36. Only small differences in the
position and intensities of the peaks were observed from one electrode
to another. According to this, none of the modifiers tested showed a
particularly intense electro-catalytic effect towards catechol.
When electrodes were covered with tyrosinase, the electrochemical

response of C-Tyr-SPE, Pt-Tyr-SPE and Au-Tyr-SPE were almost identical
to the responses observed in the absence of tyrosinase. In contrast,
GPH-Tyr-SPE, NiO-Tyr-SPE and PB-Tyr-SPE produced a shift of the anodic
peak to lower values (from 0.8 V to 0.65 V). This effect was accompanied
by a moderate increase in the intensity of the cathodic peak at -0.2 V
due to the reduction of the quinone derivative formed as a result of the
enzymatic oxidation of catechol by the immobilized tyrosinase (Figure
9.2).
217
Figure 9.2. Cyclic voltammograms registered using screen printed electrodes

modified with nickel oxide nanoparticles (red line) and nickel oxide
-1
nanoparticles-tyrosinase (blue line) immersed in 0.01 mol·L catechol in 0.01
-1 -1
mol·L phosphate buffer (pH 7.0). Scan rate 0.05V·s .
9.3.2- Measurements in musts. Optimization of the experimental

conditions
As will be shown below, voltammograms registered by immersing the
SPE electrodes in musts showed rich voltammograms where peaks
produced by components with redox activity present in the sample (e.g.
phenols, or sugars) and peaks related to the electrode modifier (e.g.
Prussian blue) could be observed.
Unfortunately, preliminary experiments using M-SPE sensors

demonstrated that the direct analysis of non-diluted musts produced
irreproducible responses. The fact that the first cycle is always different
from the rest has to be taken into account because, during this cycle, a
flux of ions diffusing inside/outside the film is established. This cycle was
always discarded. Upon successive cycling, a progressive decrease in the
intensity of the signals was observed, with a coefficient of variation (CV)
always higher than 30% (Figure 9.3.a). This was due to the existence of
proteins in suspension or the presence of other materials, such as
sugars, that could adhere to the electrode surface. Moreover, the direct
contact between the solution and the pseudo reference Ag/AgCl
electrode could also be a source of irreproducibility.
218
A 1:2 dilution with water or with 0.01 mol·L-1 phosphate buffer

drastically improved the repeatability, and the coefficients of variation
were within the range of 8-16% in M-SPE sensors not modified with
enzymes (Figure 9.3.b). The variability was lower than 10% in sensors
covered with enzymes.
Figure 9.3. Cyclic voltammograms registered using a screen printed electrode

modified with nickel oxide nanoparticles immersed in a must obtained from
grapes of Cabernet-Sauvignon variety a) undiluted sample and b) must diluted
-1
1:2 in 0.01 mol·L phosphate buffer. Graphs correspond to 10 successive cycles.
In most of the cases, a 1:2 dilution with 0.01 mol·L-1 phosphate buffer
produced voltammograms with a higher intensity than dilution with
water. However, the degree of amplification was dependent on the
nature of the electrode. This is illustrated in Figure 9.4, where a clear
increase in the intensity was observed when using GPH-Tyr-SPE whereas
the amplification was almost negligible when using a NiONP-Tyr-SPE
electrode. The reasons for the amplification are not clear, but this
phenomenon could be related to the increase in the pH produced by the
phosphate buffer.
219
Figure 9.4. Voltammetric responses of screen printed electrode modified with

a) graphene-tyrosinase and b) nickel oxide nanoparticles-tyrosinase immersed
in a must obtained from grapes of Tempranillo variety. Red lines correspond to
-1
must diluted 1:2 in water and blue lines to must diluted in 0.01 mol·L
phosphate buffer.
Regarding the lifetime, it is worth mentioning that M-SPE electrodes

decreased their intensity smoothly over 50 cycles. In contrast, M-GOx-
SPE or M-Tyr-SPE decreased their intensity progressively over ca. 10-15
cycles. After the 10-15th cycle, erratic behavior was observed and these
electrodes should be discarded. According to this, each sensor was used
to analyze one must sample.
Once the experimental conditions were established, the array of sensors

was used to analyze musts of five different grape varieties.
As mentioned in previous paragraphs the sensors developed here

provide complex voltammetric responses due to voltammetric curves
take into account every electro-active compound present in wines or
musts (e.g. polyphenols or sugars). In addition, the interaction between
the electro-catalytic material and the analytes increase the cross-
selectivity of the sensors improving the performance of the array.
Finally, the modification with enzymes and the enzyme-electron
mediator interaction introduce certain selectivity to the array.
For these reasons, each electrode showed a distinct response towards

the musts analyzed, depending on the chemical composition of the
220
working electrode. In all the M-SPE sensors, the oxidation peak of

phenols present in the musts was observed to be within the 0.5-0.8 V
region as a broad anodic wave (Figure 9.5). Different intensities were
observed depending on the nature of the working electrode.
When the electrodes were covered with GOx (Figure 9.5 blue lines), a
clear increase in the intensity of the curves was observed when the
electron mediator was NiONP or PB, confirming the excellent electron
mediator properties of nickel oxide nanoparticles and Prussian blue for
GOx. The electro-catalytic effect was not observed in C-GOx-SPE, Pt-
GOx-SPE, Au-GOx-SPE or GPH-GOx-SPE.
Figure 9.5. Voltammetric response towards must elaborated from t Tempranillo

-1
variety diluted 1:2 in 0.01 mol·L phosphate buffer (pH 7.0) of screen printed
electrodes modified with a) graphene, b) nickel oxide and c) Prussian blue. Red
lines correspond to M-CPEs response and blue lines correspond to M-GOx-CPE
-1
curves. Scan rate, 0.05 V·s .
The NiONP-Tyr-SPE and PB-Tyr-SPE biosensors also furnished

substantially larger signals than NiONP-SPE or PB-SPE analogues
reflecting the electro-catalytic activity of NiONP and PB for Tyrosinase
(Figure 9.6). This enhancement was particularly noticeable in the case of
the NiONP electrodes, demonstrating the excellent characteristics of
NiONP as electrode material in the construction of electro-chemical
enzyme sensors for the detection of phenolic compounds.
221
Figure 9.6. Voltammetric response towards must elaborated from Tempranillo

-1
variety diluted 1:2 in 0.01 mol·L phosphate buffer (pH 7.0) of screen printed
electrodes modified with a) carbon, b) nickel oxide nanoparticles and c)
Prussian blue. Red lines correspond to M-CPEs response and blue lines
-1
correspond to M-Tyr-CPE curves. Scan rate, 0.05 V·s .
As voltammetric responses depend on the chemical composition of the

studied solution and each variety of berries presents its own content in
sugars and phenols, the same sensor provided different responses
towards different musts. This is illustrated in Figure 9.7 where the
response of the PB-Tyr-SPE immersed in different musts is shown.
Figure 9.7. Cyclic voltammograms registered using a screen printed electrode

modified with Prussian Blue-tyrosinase immersed in five musts prepared from
different grape varieties.
222
9.3.3- Bioelectronic tongue. Statistical analysis

As demonstrated in the previous section, the screen-printed electrodes
were sensitive to the chemical compounds present in musts. In addition,
the chosen electron mediators showed different electro-catalytic
behaviors. According to these results, the electrodes provided distinct
responses when immersed in different musts, showing an important
degree of cross-selectivity.
Principal Component Analysis (PCA) was performed using the signals

obtained separately from the arrays of M-SPE (Figure 9.8.a), M-GOx-SPE
(Figure 9.8.b), and M-Tyr-SPE electrodes (Figure 9.8.c). Notice that each
array is formed by 6 sensors (number of variables in each case is 6
sensors10 kernels). As observed in the figure, the third principal
component also brought an important amount of information. This
result is usually found in voltammetric systems17,18,33 and is the
consequence of the large amount of information contained under the
voltammetric curves.
Figure 9.8. Principal Component Analysis scores plot using arrays of a) screen
printed electrodes, b) screen printed electrodes modified with glucose oxidase
and c) screen printed electrodes modified with tyrosinase. Must samples are 1:
Prieto Picudo; 2: Mencía; 3: Tempranillo; 4: Cabernet-Sauvignon; and 5:
Garnacha.
223
After testing the arrays separately, the multisensor systems were

merged to obtain an array that can discriminate musts according to the
sugar and phenolic content simultaneously. The combination of M-GOx-
SPE and M-Tyr-SPE considerably improved the discrimination capability.
This is illustrated in Figure 9.9.a where the scores plot, calculated when
using the combined biosensor array, is shown. As observed, using the
mixed array, a clear separation between the five musts studied was
attained. The positions of the clusters are related to sugar and phenolic
content. For instance, the Tempranillo variety which has the higher oBrix
and IPT appears apart from the rest in the negative region of PC1 and
the positive region of PC2. Prieto Picudo, Cabernet-Sauvignon and
Garnacha which have intermediate values of Brix and IPT appear in the
middle of PC1, while Mencía which presents the lowest values appears
in the negative region of PC1. It has to be noted that the cluster of the
Garnacha variety (which has similar values of IPT and Brix to Prieto
Picudo and Cabernet-Sauvignon) appears well separated from the
cluster corresponding to the other two varieties, instead of being
located in the same region. This can be attributed to the low pH values
of the must obtained from this grape which can cause important
differences in the enzymatic activity. It is also worth mentioning that
including the set of M-SPE sensors in the array did not significantly
improve the discrimination capability of the system.
Figure 9.9.b shows the loading plots obtained by using the hybrid array
formed by six M-GOx-SPE and six M-Tyr-SPE sensors (12 sensors10
kernels per sensor). As observed in the figure, biosensors have different
loadings, indicating that they provide complementary information.
The system developed showed almost the same discrimination capability

as other arrays based on more complex sensors with a high degree of
cross-selectivity.
224
Figure 9.9. Principal Component Analysis calculated using an array formed by

six screen printed electrodes modified with glucose oxidase and six screen
printed electrodes modified with tyrosinase immersed in musts of different
varieties. a) Scores plot (must samples are 1: Prieto Picudo; 2: Mencía; 3:
Tempranillo; 4: Cabernet-Sauvignon; and 5: Garnacha). b) Loadings plot
(sensors are labeled as: Graphene-Tyrosinase, Graphene-Glucose oxidase,
Platinum-Glucose oxidase, Carbon-Tyrosinase, Gold-Glucose oxidase,
Platinum-Tyrosinase, Nickel Oxide Nanoparticles-Tyrosinase, Carbon-
Glucose oxidase, Nickel Oxide Nanoparticles-Glucose oxidase, Prussian
Blue-Glucose oxidase, Gold-Tyrosinase, Prusian Blue-Tyrosinase).
The PLS model was developed in the PLS-1 mode. In order to select the
number of factors, a cross-validation method leaving out one sample at
a time, was used. Given the set of 35 calibration voltammograms, the
PLS-1 calibration on 34 calibration curves was performed and using this
calibration, the concentration of the compounds in the sample left out
during calibration was predicted. This process was repeated 35 times
until each calibration sample had been left out once. The concentration
of each sample was then predicted and compared with the known
concentration of the reference sample.
The values of the root mean square error of calibration (RMSEC), which
is an estimate of the absolute error of calibration and the values of the
root mean square of the prediction (RMSEP) and the squared correlation
coefficients (R2) obtained when plots of measured versus predicted
concentration are summarized in Table 9.3. Results indicated that the
optimized PLS-1 model allows us to assess simultaneously the IPT and
Brix in grape juices.
225
Table 9.3. Statistical parameters obtained for the Partial Least Squares
regression model established between the chemical parameters and the
a
responses of the sensors towards grapes. Root Mean Square Error of
b c
Calibration; Root Mean Square Error of Prediction. Square correlation
d e
coefficient in calibration. Square correlation coefficient in prediction; Latent
variables.
Parameter RMSECa RMESPb RC c R Pd LVe

o
Brix 0.957 5.4211 0.943 6.1827 3
Total Polyphenol Index 0.982 0.4932 0.913 1.2976 5
9.4 CONCLUSIONS
A multisensor system was obtained by using screen-printed electrodes
modified with carbon, platinum, gold, graphene, nickel oxide
nanoparticles and Prussian blue and covered with glucose oxidase or
tyrosinase. The voltammetric responses towards glucose and catechol
demonstrated that each screen-printed material shows a different
electro-catalytic and electron mediator activity, producing a variety of
responses. The electrochemical signals of musts consisted of complex
voltammetric curves bearing information associated with the phenolic
and sugar content of the grapes. In addition, the interactions between
the electrode and the solution (e.g. the pH affecting the enzymatic
activity, or the electro-catalytic activity of the modifiers) increased the
cross-selectivity of the sensors.
A clear discrimination of the samples was attained by using the hybrid

array formed by glucose oxidase and tyrosinase based sensors,
combined with PCA.
The enhanced electrochemical reactivity of the M-GOx-SPE and M-Tyr-

SPE electrodes (particularly when using Prussian Blue and nickel oxide
nanoparticles), and the variety of responses obtained when electrodes
were immersed in musts, make these modified screen-printed
electrodes attractive to form an array of sensors.
The system developed showed almost the same discrimination capability

as other arrays based on more complex sensors. Due to the limited
226
repeatability of the serigraphied electrodes, each sensor could only be

used to analyze one sample. However, the lower price, ease of use and
portability of the modified screen-printed electrode system makes the
bioelectronic tongue developed here a possible alternative tool to
analyze simultaneously sugar and phenols in situ in the vineyard block.
However, further work is needed to reduce the number of sensors, so
the system could be used more easily in a non-laboratory environment.
There is also a need to extend the study to grapes with different degree
of maturity.
227
References
1. Blouin, J., Guimberteau, G. Maturation et maturité des raisins. 2000.
Ed. Féret. Bordeaux (France).
2 Ghozlen, N.B., Cerovic, Z.G., Germain,C., Toutain, S., Latouche, G.

Sensors, 2010, 10, 10040-10068.
3. Herrera, J., Guesalaga, A., Agosin, E. Meas. Sci. Technol. 2003, 14, 689-
697.
4. Barroso, M.F., De los Santos Alvarez, N., Delerue-Matos, C., Oliveira,

M.B.P.P. Biosens. Bioelectron. 2011, 30, 1-12.
5. Carralero Sanz, V., Mena, M.L., González-Cortés, A., Yáñez-Sedeño, P.,

Pingarrón, J.M. Anal. Chim. Acta. 2005, 528, 1-8.
6. Wang, J. Chem. Rev. 2008, 108, 814-825.
7. Chaubey, A., Malhotra, B. D. Biosens. Bioelectron. 2002, 17, 441-456.
8. Ricci, F., Amine, A., Palleschi, G., Moscone, D. Biosens. Bioelectron.

2003, 18, 165-174.
9. Rahman, Md. M., Ahammad, A. J. S., Jin, J., Ahn, S.J., Lee, J.J. Sensors.
2010, 10, 4855-4886.
10. Baldwin, E.A., Bai, J., Plotto, A., Dea, S. Sensors. 2011, 11, 4744-4766.
11. Haddi, Z., Mabrouk, S., Bougrini, M., Tahri, K., Sghaier, K., Barhoumi,
H.,El Bari, N., Maaref, A., Jaffrezic-Renault, N., Bouchikhi, B. Food Chem.
2014, 150, 246-253.
12. Riul Jr., A., Dantas, C. A. R., Miyazaki, C. M., Oliveira Jr., O. N. Analyst,
2010, 135, 2481-2495.
13. Rodriguez-Mendez, M. L., Medina, C., de Saja, J.A., Apetrei, C.,

Muñoz, R. (2012). Sensor arrays based on phthalocyanines: new
developments on nanostructured and biomimetic electrochemical
228
sensors, in Multisensor systems for chemical analysis: materials and

sensors. 2012, 70-109. Ed. Pan Stanford Publishing, Inc (Singapore).
14. Sghaier, K., Barhoumi, H., Maaref, A., Siadat, M., Jaffrezic-Renault, N.
Sensor Letters. 2009, 7, 683-688.
15. Sliwinska, M., Wisniewska, P., Dymerski, T., Namiesnik, J.,

Wardencki, W. J. Agric. Food Chem. 2014, 62, 1423-1448.
16. Smyth, H., Cozzolino, D. Chem. Rev. 2013, 113, 1429-1440.
17. Apetrei, I.M., Rodríguez-Méndez, M.L., Apetrei, C., Nevares, I., del
Alamo, M., de Saja, J.A. Food Res. Int. 2012, 45, 244-249.
18. Cetó, X., Apetrei, C., Del Valle, M., Rodríguez-Méndez, M.L.
19. Cetó, X., Gutiérrez, J. M., Gutiérrez, M., Céspedes, F., Capdevila, J.,
Mínguez, S., Jiménez-Jorquera, C., del Valle, M. Anal. Chim. Acta. 2012,
732, 172-179.
20. Gay, M., Apetrei, C., Nevares, I., Del Alamo, M., Zurro, J., Prieto, N.,
de Saja, J.A., Rodriguez-Mendez, M.L. Electrochim. Acta. 2010, 55, 6782-
6788.
21. Gil-Sánchez, L., Soto, J., Martínez-Máñez, R., Garcia-Breijo, E., Ibáñez,
J., Llobet, E. Sensors Actuat. A. 2011, 171, 152-158.
22. Gutiérrez, M., Domingo, C., Vila-Planas, J., Ipatov, A., Capdevila, F.,
Demming, S., Büttgenbach, S., Llobera, A., Jiménez-Jorquera, C. Sensors
Actuat. B. 2011, 156, 695-702.

24. Rodríguez-Méndez, M.L., Apetrei, C., Gay, M., Medina-Plaza, C., De

Saja, J.A., Vidal, S., Aagaard, O., Wirth, J., Cheynier, V. Food Chem. 2014,
155, 91-97.
229
25. Zeravik, J., Hlavacek, A., Lacina, K., Skládal, P. Electroanal. 2009, 21,
2509-2520.
26. Gutiérrez, J.M., Moreno-Barón, L., Pividori, M.I., Alegret, S., del Valle,
M. Microchim. Acta, 2010, 169, 261-268.
27 Moreno i Codinachs, L., Kloock, J.P., Schöning, M. J., Baldi, A., Ipatov,
A., Bratov, A., Jiménez-Jorquera, C. Analyst, 2008, 133, 1440-1448.
28. Medina-Plaza, C., De Saja, J.A., Rodríguez-Méndez, M.L. Biosens.

Bioelectron. 2014, 57, 276-283.
29. Medina-Plaza, C., Revilla, G., Muñoz, R., Fernández-Escudero, J.A.,

Barajas, E., Medrano, G., De Saja, J.A., Rodríguez-Méndez, M.L. J.
30. Metters, J.P., Kadara, R.O., Banks, C.E. Analyst. 2011, 136, 1067-
1076.
31. OIV. Compendium of international methods of analysis of wines and

musts. Vol. 2. Organisation Internationale de la Vigne et du Vin. 2013.
Paris (France).
32. Gutiérrez-Osuna, R., Nagle, H.T. IEEE Trans. Syst. Man. Cybern. B-
Cybern. 1999, 29, 626-632.
33. Prieto, N., Oliveri, P., Leardi, R., Gay, M., Apetrei, C., Rodriguez-
Mendez, M.L., de Saja, J.A. Sensors Actuat. B. 2013, 183, 52-57.
34. To Thi Kim, L., Gabrielli, C., Perrot, H., Garcia-Jareno, J., Vicente, F.
35. Valentini, F., Carbone, M., Palleschi, G. Anal. Bioanal. Chem. 2013,
405, 3449-3474.
36. Matemadombo, F., Apetrei, C., Nyokong, T., Rodríguez-Méndez,

M.L., de Saja, J. A. Sensors Actuat. B. 2012, 166-167, 457-466.
230
CHAPTER 10
Nanoscale Au-In Alloy-Oxide Core-Shell Particles
as Electrocatalysts for Efficient Hydroquinone
Detection
Despite the health benefits and antioxidant properties present on

phenols, they can also occur as a hazardous byproduct of different
industries.
Hydroquinone (HQ) is used as a developing agent in black-and-white

photography, lithography and x-ray films. It is also used as an
intermediate to produce antioxidants for rubber. It is added to a number
of industrial monomers to inhibit polymerization during shipping,
storage, and processing. HQ is on the Hazardous Substance List because
it is regulated by Occupational Safety and Health Administration (OSHA).
This chapter was develop in the collaboration with Dr. Eli Sutter at
Brookhaven National Lab (NY,USA). The group of Dr. Eli Sutter focuses its
research on new nanomaterials with energy and environmental
applications. This collaboration led to new voltammetric sensors applied
to detect HQ with environmental purposes.
For this goal, voltammetric sensors of high-surface area AuxIn1-x core-

shell nanostructures will be studied. AuxIn1-x with different alloy
composition, x, were prepared by room temperature oxidation of Au-In
alloys. Au-In core-shell nanoparticles have been characterized by XPS,
EDS, TEM and SEM and used to modify standard indium tin oxide (ITO)
electrodes and investigate the electrocatalytic properties towards the
oxidation of hydroquinone in aqueous solutions. The electrocatalytic
effect, the influence of Au:In ratio on the electrochemical analysis and
the dynamic behavior of the sensors will be discussed. This research has
been published on Journal of Physical Chemistry-C.
10.1 INTRODUCTION
Phenolic compounds are a large family of molecules, which have
attracted great interest in the food industry due to their antioxidant
capacity and potential health benefits1 .On the negative side, phenolic
compounds occur as byproducts of a variety of industries and are
ubiquitous in the environment. They are classified as hazardous
materials and are particularly toxic to the environment. Hydroquinone
(HQ) is one example of this broader class of phenolic compounds that is
233
important in a number of biological processes, as well as in industrial

products including dyes, cosmetics, pesticides, paper manufacturing,
and photographic developers2 .The detection and assessment of
hydroquinone can be carried out using traditional techniques such as
fluorescence3 ,chemiluminescence4 and high performance liquid
chromatography (HPLC)5 among others. Compared to these methods
that require complex instrumentation, electrochemical methods could
provide compact and relatively inexpensive setups with fast response for
real-time analysis in the field. Conventional (e.g. metallic or glassy
carbon) electrodes however, are not suitable for the detection of
phenols due to their poor electrochemical response6. One possible
strategy for preparing efficient sensing devices is the modification of the
electrode surface with nanomaterials. Electrodes modified with metal
nanoparticles are very promising as they show unique electronic and
catalytic properties7. Au nanoparticles, among other noble metals, have
demonstrated electrocatalytic enhancement in various processes: fuel
cells8,9, sensors10,11, food analysis12,13 and photovoltaic devices14,15. Au
nanoclusters provide even higher catalytic activity16 as well as other
benefits, for instance stabilization against dissolution under potential
cycling regimes of oxygen-reduction fuel-cell electrocatalysts17. Small Au
nanoparticles and nanoclusters require complicated synthesis and are
hard to stabilize in their active configuration, as they sinter easily and
their catalytic properties degrade as a result of the size increase.
Recently, a simple approach for the formation of stable Au-based
catalysts, in which the active sites are small gold clusters anchored in an
amorphous Au-In oxide matrix, has been developed. Au-In alloy
nanoparticles prepared by metal evaporation (or other methods)
transform into metal alloy-oxide core-shell structures during oxidation at
room temperature18. AuIn alloy nanoparticles terminated by gold-
containing oxide shells were found to be active in the conversion of CO
and O2 to CO2 near room temperature, as well as stable against sintering
up to high temperatures (>300oC), and thus show promising properties
for heterogeneous catalysis.
234
10.2.1- Synthesis of AuxIn1-x alloy nanoparticles

Bimetallic Au-In alloy nanoparticles with different compositions were
formed by sequential room-temperature magnetron sputtering of
controlled amounts of Au and In onto different substrates: indium tin
oxide (ITO) electrodes (for electrocatalytic measurements), as well as Ge
(111), amorphous carbon and few-layer graphene membranes
supported on standard transmission electron microscopy (TEM) grids for
materials characterization. Pure In and Au nanoparticles for control
experiments were prepared by sputtering of only In and Au,
respectively, under the same conditions. Different compositions of the
Au-In alloy in the nanostructures were obtained by depositing a fixed
amount of Au (2nm equivalent thickness), followed by sputtering of
different amounts of In to obtain the desired composition. The as-
deposited nanoparticle ensembles were exposed to ambient conditions
after removal from the growth reactor, which led to their oxidation in air
and formation of mixed oxide shells encapsulating the bimetallic alloy
cores18-20.
10.2.2- Characterization of Au-In alloy nanoparticles

The morphology and composition of the nanoparticles were investigated
by TEM in a JEOL 2100F field-emission microscope equipped with energy
dispersive X-ray spectroscopy (EDS) instrument, in scanning TEM mode
(beam size: 2Å). Laboratory X-ray photoelectron spectroscopy (XPS)
spectra on nanoparticle ensembles deposited on Ge(111) were acquired
at room temperature using a Specs Phoibos 100 MCD hemispherical
analyzer, using excitation by Al Kα (hν=1486.6 eV) radiation at 300 W (10
keV; 30 mA). XPS spectra were acquired with pass energy of 25 eV and
energy step of 0.05 eV. The base pressure in the analysis chamber was in
the range of (2-6)10-9 Torr. The C 1s binding energy (284.8 eV) was
used as binding energy reference. To determine the composition in the
outermost surface layer, low-energy ion-scattering spectroscopy (ISS)
measurements were carried out in a ultrahigh vacuum (UHV) system
235
with (1-3)10-10 Torr base pressure, using 2 keV He ions (Specs IQE 12/38
ion source).
10.2.3- Electrochemical characterization

Cyclic voltammetry (CV) of ITO electrodes modified with Au-In alloy-
oxide core-shell nanoparticles, prepared as described above, was carried
out in an Autolab PGSTAT128N potentiostat-galvanostat using a three-
electrode cell. Ag|AgCl/KCl 3M was used as reference electrode and the
counter electrode was a platinum plate. ITO glasses covered with Au-In
nanostructures with different compositions were used as working
electrode. Experiments were carried out in the bias range from -0.8 to
1.0 V using KCl 0.1 mol·L-1 as supporting electrolyte in which
hydroquinone (HQ) was dissolved to obtain a final concentration of 10-3
mol·L-1. The scan rate used in CV was 0.1 V·s-1, with the influence of the
scan rate analyzed in the range from 0.025 to 0.5 V·s-1.
10.3.1- Structural characterization

Figure 10.1 summarizes the structure and morphology of
nanostructured AuxIn1-x alloys with different compositions, produced by
sequential vacuum deposition of Au and In, as well as pure Au
nanoparticles. The TEM images showed that magnetron sputtering
yielded discrete Au nanoparticles with a large variety of sizes (Figure
10.1.a), while the typical morphology obtained for a wide composition
range Au-In bimetallic alloys following room-temperature oxidation in
air (Figure 10.1.b-d) are elongated, meandering nanostructures and
nanoparticles exhibiting a high-surface area morphology. The TEM
images also showed the partial dewetting of the Au-In alloy into stripes,
between which the amorphous C support is exposed. High resolution
TEM (Figure 10.1.b-d, insets) showed that the meandering
nanostructures consist of a crystalline core with darker contrast and
interspersed areas with brighter contrast that stem from the amorphous
oxide shells, similar to earlier observations18. Scanning electron
microscopy (SEM) investigation on ITO electrodes and crystalline Ge
substrates confirmed that the Au-In nanostructures have the same
236
morphology, independent of the support. The surface coverage

increased with the increase of In content, both because the amount of
deposited metal was larger and because the resulting oxides were
thicker (Figures 10.2.b and 10.2.c). Experimental electron diffraction
patterns for the four different samples (Figure 10.1.a’-d’) were
compared to simulated diffraction patterns for Au, orthorhombic
Au3In21, and cubic AuIn222 structures and confirmed the crystal structure
and composition of the cores. Crystallographic studies of bulk Au0.5In0.5
found that it adopts a pseudo-orthorhombic structure23, however, the
atomic positions have not been reported to date, so we could not
simulate the diffraction pattern (DP) for this composition.
Figure 10.1. Overview TEM images of the characteristic structure of a) Au and

Au−In samples with different compositions, b) Au3In, c) AuIn and d) AuIn2
prepared by magnetron sputtering. High-resolution TEM images showing details
of the crystalline Au nanoparticles and Au−In nanostructures terminated by
amorphous surface oxides are shown in the insets of figures a−d. a′−d′ Electron
diffraction patterns from the nanostructures in figures a−d taken with a 150 nm
aperture. The experimental diffraction patterns shown in the left half are
compared to simulated diffraction patterns (right half) calculated using the
21
software package JEMS for nanoparticles with Au, orthorhombic Au 3In and
22
cubic AuIn2 structures.
237
The Au0.5In0.5 nanostructures were identified unambiguously from the

EDS and XPS measurements (Figures 10.2 and 10.3). It is worth noting
that the structures with Au0.5In0.5 composition showed distinct DP from
the DPs of other alloys, which confirmed that Au0.5In0.5 represents a
distinct crystalline phase and can be used for the identification of this
alloy as well.
EDS measurements were used to confirm the different compositions of

the nanostructures depending on the amounts of Au and In deposited.
Figure 10.2.a shows characteristic EDS spectra from Au3In, AuIn, and
AuIn2 nanostructures. The spectra confirmed the alloying of the
sequentially deposited metals at room temperature18,24. Following air
exposure, amorphous oxide shells that contain both Au and In covered
the surface of the Au-In alloy nanostructures (Figure 10.2.b). Analysis of
the TEM images (Figure 10.2.c) showed that the thickness of the oxide
shells depends on the composition of the initial Au-In alloy: 1.0±0.4 nm
for Au3In, 1.8±0.8 nm for AuIn, and 2.8±1.2 nm for AuIn2 nanostructures.
This trend in the oxidation rate with alloy composition is the same as the
one established for discrete Au-In particles18. The amorphous mixed
oxide forming on the surface of the particles during room temperature
oxidation in air scales with In content, i.e. the oxide shells are thicker for
In-rich nanoparticles.
Figure 10.2. a) EDS spectra from Au−In nanostructures with different alloy
compositions (Au3In, AuIn, AuIn2). b) High-resolution TEM image showing detail
of the crystalline AuIn nanostructures terminated by an amorphous surface
oxide after exposure to air at room temperature. c) Thickness distribution of the
amorphous oxide shells formed after oxidation in air for 28 days for three
different as-deposited alloy compositions.
238
Core-shell AuxIn1-x oxide nanostructures were further investigated using

XPS to determine the composition and chemical environment, and ISS to
identify the elements in the outermost layer of the oxide shell. Both
investigations were performed on oxidized nanostructures on Ge
support. XPS (in conjunction with TEM) was used to monitor the
thickening of the oxide on the surface until the oxide reached its limiting
thickness, identified by the absence of further changes in the XPS
spectra. For the Au-In alloys investigated here, this limit was reached
after ~21 days in air. XPS on all Au3In, AuIn and AuIn2 nanostructures
revealed the presence of Au, In, and O. The Au 4f and In 3d XPS spectra
for these three alloy compositions together with reference spectra for a
2 nm Au film and pure In2O3 nanoparticles (on Ge) are shown in Figures
10.3.a and 10.3.b, respectively. As expected, the Au 4f spectra of the
oxidized alloys were shifted to higher binding energy compared to the
Au reference and the observed shifts increased with increasing In
content, in good agreement with previous results on Au-In alloys25. The
In 3d peaks for the three alloys were also shifted compared to the
spectrum of pure In2O3 nanoparticles.
a) b)
Figure 10.3. a) X-ray photoelectron spectra of the Au 4f core levels of the

composite Au-In-amorphous oxide core-shell nanostructures compared to a
reference spectrum of Au nanoparticles. b) X-ray photoelectron spectra of the
In 3d core levels from the nanostructures compared to a reference sample of
In2O3 nanoparticles on Ge substrate.
239
Because of the high energy of the exciting photons from a laboratory

XPS source, the kinetic energy of the Au 4f and In 3d photoelectrons is
high, which translates into a large escape depth. As a result the XPS
measurements probed both the alloy core and the mixed oxide shell of
the nanostructures and were of limited use for characterizing the near-
surface composition of the Au-In oxide structures, which is key to their
electrocatalytic properties. Thus, ISS was used to determine the
constituents of the surface of oxidized Au-In alloy nanoparticles (Figure
10.4.c-e), with a bulk Ge crystal (with native oxide, Figure 10.4.a), Au
film on Ge (Figure 10.4.b), and In2O3 nanoparticles on Ge (Figure 10.4.f)
serving as reference samples.
ISS on the nanostructured Au-In oxide samples showed the presence of

Au, In, Ge, and O in the surface layer (Figure 10.4.c-e). This confirmed
that all oxidized Au-In nanostructures contain Au at the surface of their
oxide shells. The Ge signal stemed from the exposed substrate between
the nanoparticles. In the alloy nanostructures with higher In content (i.e.
larger overall coverage of deposited metal) the Ge peak was not well
resolved because of the small remaining area of the bare support in
these samples (Figure 10.1.d). These samples have the thickest oxide
shells (Figure 10.2.c), which contribute to the almost complete shielding
of the substrate from the incoming He ions. The (Au:In) peak ratios in
samples with different In content showed a clear trend in Au content,
which follow the composition of the initial Au-In alloy: the oxide surface
of particles with Au3In cores contains significantly more Au (Au:In = 0.79)
than that of particles with AuIn cores (Au:In = 0.22), which again is larger
than that of particles with AuIn2 cores (Au:In = 0.11). From the ISS
measurements, it can be concluded that the outermost atomic layer of
the amorphous oxide shells contains Au, In and O. The amount of Au
exposed at the oxide surface is not fixed, but can be varied by adjusting
the Au concentration in the initial Au-In binary alloy. This provided
access to working electrodes with systematically different
concentrations of near-surface Au, stabilized on the amorphous Au-In
oxide shells, for electrochemical characterization.
240
a)
b)
c)
d)
e)
f)
Figure 10.4. Ion scattering spectra of: a) Reference Ge wafer, used as a

substrate for the nanostructures, b) Reference Au film, c) Au3In, d) AuIn, e)
AuIn2-amorphous Au-In oxide core-shell nanostructures and f) crystalline In2O3
nanoparticles.
10.3.2- Electrocatalytic properties

The electrocatalytic properties of the Au-In alloy nanostructures for the
detection of HQ were measured in aqueous solutions containing 10-3
241
mol·L-1 HQ. This particular concentration was chosen to facilitate

comparison with other means of detection, as it is the one usually
targeted for HQ detection. Cyclic voltammograms were measured using
ITO electrodes modified with the core-shell nanostructures with
different (Au:In) ratios. Reference voltammograms were recorded for
unmodified ITO and bulk Au electrodes, as well as working electrodes
modified with Au nanoparticles. According to the literature, in aqueous
solution with neutral pH and using carbon electrodes, para-
hydroquinones undergo a two-electron single-step oxidation26 that
appears as a broad and irreversible anodic peak in the 0.7–1.0 V region.
The characteristic voltammograms registered using unmodified ITO-on-
glass electrodes, bulk gold and Au nanoparticle-modified ITO electrodes
are compared in Figure 10.5.a. The voltammograms measured with
unmodified ITO and ITO modified with the Au-In nanostructures as the
working electrode are shown in Figure 10.5.b. In Figure 10.5, all
voltammograms are shown on the same scale to allow direct
comparison of the currents. In Figure 10.6, the voltammograms
registered with the different electrodes are shown on different scales, so
that their details can be clearly seen. The characteristic voltammogram
registered using unmodified ITO the working electrode, is shown in
Figures 10.5.a (black curve) and 10.6.a. Starting at 0.75 V, a small
increase in the current was observed, indicating that with an unmodified
ITO working electrode HQ oxidation starts at high voltages, probably due
to the difficulty of the electron transfer to the ITO glass. Thus,
unmodified ITO electrodes are not suitable for the detection of HQ.
242
Figure 10.5. Cyclic voltammograms of a) ITO (black), bulk Au electrodes (dark

blue) and Au nanoparticle modified ITO electrodes (yellow curve) and b) ITO
(black) and electrodes modified with AuIn2 (blue), AuIn (red) and Au3In (orange)
core-amorphous mixed oxide shell nanostructures immersed in an aqueous HQ
-3 -1 -1
solution (10 mol·L ). Scan rate, 0.1 V·s .
The voltammogram recorded with a bulk gold working electrode is

shown in Figure 10.5.a (dark blue curve) and 10.6.b. The entire curve
presented very low intensity. Despite the low overall intensity, the
voltammogram (Figure 10.6.b) showed a well-defined redox pair at ~0.7
V and ~0.2 V (labeled Ia and Ic), respectively. This redox process is
associated with the two-electron (reduction-) oxidation process of the
quinone-hydroquinone couple in buffered aqueous media26 that can be
written as: QH2  Q + 2H+ + 2e-.
Scheme 10.1. Two-electron two-proton reduction of quinone in aqueous buffer.
243
For the bulk Au electrode, several additional peaks were observed as

well. The most important is the redox pair III, which is related to a weak
dimerization process of the quinoid form. Notice that peak IIIa can be
clearly observed, whereas peak IIIc overlaps with peak Ic. In addition,
several cathodic peaks arising from redox processes associated with
adsorbates on the gold surface (usually –OH of the solvent linked to the
gold surface) were detected27. Importantly the current densities
recorded with the bulk Au electrode were very low, rendering it almost
inert and not suitable to efficiently detect the oxidation of HQ.
The coverage of the ITO/glass electrodes with nanoparticles induced

new reaction mechanisms leading to significant changes in the cyclic
voltammograms. The voltammograms registered using ITO/glass
electrodes with Au nanoparticles with broad distribution of sizes in the
range between 1 and 20 nm, (Figures 10.5.a (yellow curve) and 10.6.c),
were characterized by a drastic increase in intensity. In addition, the
presence of gold nanoparticles changes the mechanism of oxidation and
the HQ molecules undergo a two-step one-electron oxidation. This led to
a new peak (IIa) which was clearly observed at ~0.0 V corresponding to
the first one-electron oxidation, while the second one-electron oxidation
corresponded to peak Ia. In this case, the number of dimers was higher
justifying the change in the relative intensities Ia/IIIa. Other cathodic
peaks observed were related to the redox process associated with
adsorbates on the gold surface, similar to the bulk Au electrode. The
current densities were large, thus, there is high electrocatalytic activity
stemming from the Au nanoparticles. However, the CV was quite
complicated due to the activation of several processes such as HQ
oxidation, dimerization, etc.
Results for ITO working electrodes modified with Au-In nanostructures

with bimetallic core and oxide shell are discussed in the following
paragraphs. A characteristic voltammogram recorded with a working
electrode modified by oxidized AuIn2 is shown in Figures 10.5.b (blue
line) and 10.d. The voltammogram was rather simple. The onset of the
244
oxidation peak was shifted by ~0.3 V to lower potential compared to the

ITO electrode, and the increase of the current was much larger.
The changes in the voltammograms were even more pronounced for ITO
electrodes modified with AuIn core-oxide shell nanoparticles (Figures
10.5.b and 10.6.e). Here the onset of oxidation was again shifted to
lower potentials, similar to the AuIn2 modified electrode. In addition, a
well-defined peak with an onset at 0.25 V and maximum at ~0.55 V was
observed. This behavior is consistent with the two-electron oxidation of
quinone-hydroquinone couples in buffered aqueous medium28.
Figure 10.6. Cyclic voltammograms of a)ITO, b) bulk Au electrodes and

electrodes modified with c) Au nanoparticles, d) AuIn2, e) AuIn and f) Au3In core-
-3 -1
amorphous-mixed-oxide-shell nanostructures immersed in HQ (10 mol·L ).
-1
Scan rate, 0.1 V·s .
245
On the extreme Au-rich side, finally, ITO electrodes modified with

oxidized Au3In core-shell nanoparticles produced a fundamentally
different response. Two separate anodic waves Ia and IIa accompanied by
their corresponding reduction peaks Ic and IIc were observed (Figures
10.5.b and 10.6.f). This behavior is consistent with two successive one-
electron oxidation steps (Scheme 10.2) in which the first step is
completely reversible, leading to the formation of semiquinone, and the
second quasi-reversible step gives rise to the quinone. Such a behavior
has previously been reported in aprotic solvents29, but this is the first
time that this two-step oxidation process has been found in aqueous
media.
Scheme 10.2. Two-electron two-step one-electron oxidation of quinone in

aqueous buffer.
From the above results, it can be concluded that Au-In core-shell

nanoparticles showed clear electrocatalytic effects that depend on the
Au/In ratio of the initial alloy, which in turn determines the amount of
Au stabilized in the near surface area of the oxide on the surface of
these nanostructures. ITO working electrodes modified with
nanoparticles with moderate Au content (AuIn2 and AuIn) caused a shift
of the oxidation peak to lower potentials (i.e. facilitated the oxidation)
and showed much higher activity than unmodified ITO electrodes.
Importantly for the detection of HQ, the electrocatalytic (reduction-)
oxidation of HQ on these electrodes follows a single pathway, the two-
electron oxidation of the quinone-hydroquinone couple. This contrasts
with the behavior of Au nanoparticle modified ITO electrodes, which
also showed large current densities (i.e. high activity) but were not
selective to a single reaction mechanism and instead showed a number
246
of concurrent processes. Au-rich Au3In nanoparticle electrodes not only

facilitated the oxidation but also provided a different pathway, a two-
steps mechanism involving two successive one-electron oxidation steps,
to the oxidation of HQ. In general, dimerization and polymerization are
avoided when Au-In nanoparticles are used as working electrodes,
showing more stables responses and clearer curves. Finally, it is worth to
point that Au/In ratio influenced the intensity of the responses. The
intensity of the oxidation peaks was higher when the mechanism was a
two-electron process, and the intensity was much higher at low Au
content.
10.3.3- Dynamic behavior

The influence of the scan rate on the peak height was also investigated
in the 0.025-0.500 V·s-1 range. The redox peak currents increased
progressively with the scan rate (Figure 10.7.a). The intensity of the peak
current (measured at the voltage of oxidation or reduction of HQ) scaled
with the square root of the scan rate (Figure 10.7.b), which indicates a
diffusion controlled mechanism for the oxidation/reduction of the
antioxidant. This indicates that the high surface area of the
nanostructured electrode material facilitates a rapid electron transfer
between electrode surface and the analyte.
-3 -1
Figure 10.7. a) CVs obtained for HQ (10 mol·L ) at Au3In modified electrode at
-1
different scan rates (0.025-0.5 V·s ). b) Plot of peak current, measured at the
voltage of oxidation () or reduction () of hydroquinone, as a function of (scan
1/2
rate) .
247
The slopes obtained of the peak current versus (scan rate)1/2

characteristics for the oxidation peak were different for the different
alloys (Table 10.1). A higher value of the slope indicates faster charge
transfer. The highest value was found for AuIn nanoparticles, indicating
that AuIn modified working electrodes were most effective in facilitating
the electrode-to-analyte charge transfer.
Table 10.1. Slope and regression coefficient of linear fits of the intensity of the
anodic and cathodic peaks of the hydroquinone redox process as a function of
-1
the square root of the scan rate (between 0.025 – 0.5 V·s ).
Slope (μA(mV·s)-1/2) R2
Sensor
Oxidation Reduction Oxidation Reduction
AuIn2 2.4 - 0.993 -
AuIn 55.0 - 0.994 -
Au3In 13.4 9.4 0.992 0.998
10.3.4- Limits of detection

The effect of the concentration of HQ in the response of the sensors was
investigated in the range 10-5 to 10-3 mol·L-1. The sensitivity was obtained
from the slope of the graph. The corresponding Limits of Detection
(LOD) were calculated according to the 3sd/m criterion where m was the
slope of the calibration graph, and sd was estimated as the standard
deviation (n = 5) of the signals at the concentration level corresponding
to the lowest concentration of the calibration plot. Results are shown in
Table 10.2.
Table 10.2. Sensitivity and LOD towards HQ for ITO working electrodes modified
with different oxidized Au-In nanostructures, compared to AuNP and Au bulk
electrodes. Relative standard deviation: 1.8% (Au 3In) and 7-8% (AuIn, Au NPs,
bulk Au)
AuIn2 AuIn Au3In AuNPS Au bulk

Sensitivity - 0.201 0.075 0.487 0.005
LOD (mol·L-1) - 1.3310-5 3.5710-5 5.4910-6 5.0510-4
248
The LODs decreased when increasing the amount of Au exposed on the

amorphous oxide surface. The LODs were determined to be in the range
of 10-5-10-6 mol·L-1 and were lower than those obtained using bulk Au.
However, it is important to notice that AuIn2 cores did not show good
electrocatalytic properties, and peaks associated to HQ shifted to even
higher potentials when increasing the concentration of antioxidant. For
this reason, the limit of detection could not be calculated accurately for
AuIn2 films.
To establish the reproducibility of the measurements with the different

electrodes the determination of 10-3 mol·L-1 HQ was repeated six times.
The first cycle was always different from the rest of the scans and was
discarded. This behavior is typically observed in chemically modified
electrodes and is due to the diffusion of ions inside/outside the films. In
the case of Au3In, subsequent cycles achieved a good reproducibility
with a relative standard deviation (RSD) of 1.8%. In the rest of the films
studied the intensity of the signals decreased progressively and the RSD
calculated was in the range of 7-8%.
10.4 CONCLUSIONS
Oxidized Au-In core-shell nanoparticles with different compositions were
investigated as electrocatalysts for the (reduction-) oxidation of HQ. The
nanoparticles were characterized by TEM, XPS and ISS measurements
and these combined measurements demonstrated that the amorphous
mixed oxide shells formed on the Au-In nanoparticles with different alloy
compositions had different amount of Au and In on the surface. ITO
electrodes modified with oxidized Au-In core-shell nanostructures
present different electrocatalytic activity depending on the amount of
Au stabilized on their surfaces. Nanoparticles with AuIn2 and AuIn core
surrounded by oxide shells produced a displacement of the onset of the
peak to lower potentials, whereas Au3In samples caused two separate
anodic peaks with their corresponding cathodic waves. It can be
concluded that among the investigated alloy compositions the oxidized
AuIn core-shell nanoparticles promise the best performance for the
analysis of antioxidants due to the excellent range where the redox
249
peaks are detected, the excellent definition of the redox peaks with high
intensity, and enhancement of the charge transfer.
250
References
1. Vauzour, D., Rodriguez-Mateos, A., Corona, G., Oruna-Concha, M.J.,
Spencer, J.P.E. Nutrients. 2010, 2, 1106-1131.
2. Zhang, F., Li, M.A., Li, W.Q., Feng, C.P., Jin, Y.X., Guo, X., Cui, J.G.
Chem. Eng. J. 2011, 175, 349-355.
3. Pistonesi, M.F., Di Nezio, M.S., Centurion, M., Palomeque, M.E., Lista,

A.G., Band, B.S.F. Talanta. 2006, 69, 1265-1268.
4. Cui, H., Zhang, Q.L., Myint, A., Ge, X.W., Liu, L.J. J. Photoch. Photobio.
A. 2006, 181, 238-245.
5. Liu, C., Wang, J., Yang, Y.L. Anal. Methods-UK. 2014, 6, 6038-6043.
6. Sakthinathan, S., Palanisamy, S., Chen, S.M., Wu, P.S., Yao, L., Lou, B.S.
Int. J. Electrochem. Sci. 2015, 10, 3319-3328.
259.
8. Kuralkar, M., Ingle, A., Gaikwad, S., Gade, A., Rai, M. IET
Nanobiotechnology [Online], 2015, 66-70. http://digital-
library.theiet.org/content/journals/10.1049/iet-nbt.2014.0004.
9. Kilic, M.S., Korkut, S., Hazer, B., Erhan, E. Biosens. Bioelectron. 2014,
61, 500-505.
10. Zhao, Z., Zhang, M., Li, Y., Cheng, S., Chen, X., Wang, J. Anal. Lett.
2015, 48, 1437-1453.
11. Medina-Plaza, C., Furini, L.N., Constantino, C.J.L., de Saja, J.A.,

12. Lin, X., Ni, Y., Li, S., Kokot, S. Analyst. 2012, 137, 2086-2094.
13. Medina-Plaza, C., García-Cabezón, C., García-Hernández, C.,

Bramorski, C., Blanco-Val, Y., Martín-Pedrosa, F., Kawai, T., de Saja, J. A.,
Rodríguez-Méndez, M. L. Anal. Chim. Acta. 2015, 853, 572-578.
251
14. Chuang, M.K., Lin, S.W., Chen, F.C., Chu, C.W., Hsu, C.S. Nanoscale.
2014, 6, 1573-1579.
15. Vijayakumar, C., Balan, B., Saeki, A., Tsuda, T., Kuwabata, S., Seki, S. J.
Phys. Chem. C. 2012, 116, 17343-17350.
16. Corma, A., Concepción, P., Boronat, M., Sabater, M. J., Navas, J.,
Yacaman, M.J., Larios, E., Posadas, A., López-Quintela, M.A., Buceta, D.,
Mendoza, E., Guilera, G., Mayoral, A. Nat. Chem. 2013, 5, 775-781.
17. Zhang, J., Sasaki, K., Sutter, E., Adzic, R.R. Science. 2007, 315, 220-
222.
18. Sutter, E.A., Tong, X., Jungjohann, K., Sutter, P.W. P. Natl. Acad. Sci.
USA. 2013, 110, 10519-10524.
19. Sutter, E., Sutter, P. J. Phys. Chem. C. 2012, 116, 20574-20578.
20. Robinson, M.C., Slavin, A.J. Phys. Rev. B. 1996, 54, 14087-14092.
21. Jayne, D.T., Fatemi, N.S., Weizer, V.G. J. Vac. Sci. Tech. A 1991, 9,
1410-1415.
22. Bailey, S.I., Ritchie, I.M. Electrochim. Acta. 1985, 30, 3-12.
23. Deplanche, K., Merroun, M.L., Casadesus, M., Tran, D.T., Mikheenko,
I.P., Bennett, J.A., Zhu, J., Jones, I.P., Attard, G.A., Wood, J., Selenska-
Pobell, S., Macaskie, L.E. J. R. Soc. Interface. 2012, 9, 1705-1712.
24. Guin, P.S., Das, S., Mandal, P.C. Int. J. Electrochem. 2011, 2011, 22.
25. Gupta, N., Linschitz, H. J. Am. Chem. Soc. 1997, 119, 6384-6391.
252
CHAPTER 11
Final Conclusions
Final Conclusions
According to the objectives stablished and the results obtained, the

conclusions reached during the elaboration of this thesis are the
following:
1- Voltammetric sensors and biosensors have been desingned and

developed using different techniques and electrocatalytic materials.
They have been successfully used to the analysis of diverse analytes
present in musts such antioxidants (catechol, hydroquinone…), organic
acids and sugars.
a) Functionalized (n-dodecanethiol) gold nanoparticles sensors

were successfully developed by Langmuir-Blodgett technique. The
modified electrodes showed an enhancement in the sensitivity towards
organic and phenolic acids (malic, tartaric, lactic, gallic and cafeic), an
increase in the current and a shift of the potential in the less positive
direction, compared to an ITO bare electrode. Efficient catalytic activity
was also observed in mixtures of tartaric/caffeic acid, being the
electrode able to provide simultaneous information about the
concentrations of both analytes without any interference.
b) Lutetium Bis-Octachloro-Phthalocyanate sensors were

developed by Langmuir Blodgett and Lagmuir-Schaefer techniques. The
nanostructure of the films is responsible of the improved
electrochemical behavior towards the analysis of catechol when
compared with cast film. The sensors presented good stability and
sensitivity reaching limits of detection around 10-5 mol·L-1.
c) Two electrocatalytic materials (functionalized gold

nanoparticles and lutetium bisphthalocyanine) were deposited onto ITO
glass using Langmuir-Blodgett technique. Synergistic electrocatalytic
effect was observed towards the analysis of hydroquinone, getting an
increase in the current and limits of detection in the order of 10-7 mol·
L-1.
d) Carbon paste electrodes of dilithium phthalocyanine were

prepared using different carbonaceous materials (graphite, carbon
255
Chapter 11
microspheres and multiwall carbon nanotubes). The electrochemical

response of the electrode was found to be dependent on the nature of
the matrix. The electrodes were used to analyze citric acid facilitating
the oxidation of the analyte. Synergistic effect was also found when
MWCNT and LiPc2 were used, improving the electrocatalytic activity.
e) Au-In alloy-oxide core-shell particles prepared by magnetron

sputtering showed electrocatalitic effect towards the detection of
hydroquinone. The electrocatalytic activity and the mechanism of the
redox process depended on the amount of Au stabilized on the surface.
Limits of detection in the order of 10-5-10-6 mol·L-1 were reached.
2- Arrays of sensors and biosensors have been applied to the

discrimination of musts from different varities of grapes. In order to get
information about the behavior of the electrodes, model solutions were
analyzed previously to the samples.
a) An electronic tongue formed by sensors and biosensors

containing different metallophthalocyanines and enzymes (tyrosinase
and glucose oxidase) was developed using carbon paste technique. The
response was found to be dependent on the nature of the
phthalocyanine and improved by the presence of the enzymes. The array
was applied to analyze musts of different varieties of red grapes and its
capability of discrimination was evidenced using Principal Component
Analysis. The selectivity of the multisensor system and its capability of
discrimination were clearly improved when biosensors were included in
the array.
b) Commercial screen-printed electrodes were modified with

enzymes (tyrosinase and glucose oxidase) in order to obtain a
multisensory system devoted to the discrimination of red grapes
varieties. The sensors presented different electron mediators: carbon,
platinum, Prussian blue, gold, graphene and nickel oxide nanoparticles.
The hybrid array was able to discriminate samples from different
varieties of grapes as shown by Principal Component Analysis. The
results obtained presented good correlation with Brix and Total
256
Final Conclusions
Polyphenol Index analyzed by traditional methods. The advantages of

this hybrid electronic tongue are the lower price of the sensors, easy use
and portability of the system.
c) A bioelectronic tongue formed by nanostructured biosensors

containing phenol oxidases (laccase and tyrosinase) was developed using
Langmuir-Blodgett technique. Lutetium bisphthalocyanine was used as
electron mediator and arachidic acid as a lipid in order to mimic cell
membranes. Biomimetic structure improved enzymes performance
allowing getting limits of detection around 10-8 mol·L-1. The array was
able to discriminate model solutions of phenols according to the number
of phenolic groups present on their structure and musts from different
varieties of grapes according to their total polyphenolic content as
demonstrated by Principal Component Analysis.
3- A bioelectronic tongue containing enzymes devoted to the

detection of phenols (tyrosinase and laccase) and sugars (glucose
oxidase and D-fructose dehydrogenase) was employed to monitor the
ripening process of the grapes.
a) Enzymes were imbibed in a biomimetic environment using

Langmuir-Blodgett technique to increase their performance. Lutetium
bisphtalocyanine was employed as electron mediator to improve the
electron transfer within the electrode.
b) Musts from different varieties of red grapes were

discriminated using the bioarray of sensors. The scores plot obtained
using Principal Component Analysis places the must on the basis of the
phenolic content and sugar concentration. In addition, when samples of
the same variety and harvested in different vintages were analyzed the
system was able to discriminate them according to the variety and the
vintage.
c) The bioelectronic tongue was used to analyze grapes

harvested from the veraison to over-ripening in a weekly basis. Principal
Component Analysis showed a clear discrimination of the samples
257
Chapter 11
collected during the progression of ripening being able to follow the

maturing process.
d) The results obtained using the array of biosensors were

correlated to the traditional chemical parameters by means of Partial
Least Square-1 obtaining excellent correlations for all parameters
related to sugar and phenolic content.
258
RESUMEN EN ESPANOL
Resumen en Español
1- INTRODUCCIÓN
El sector de agroalimentación constituye una de las bases más
importantes de la economía de Castilla y León, originando un cuarto de
la producción industrial de la región. La industria enológica es uno de los
sectores más tradicionales e importantes de Castilla y León donde el
vino de calidad es considerado patrimonio cultural y gastronómico.
En los últimos años el sector enológico has sido confirmado como una de
las partes más dinámicas en el sector agroalimentario en Castilla y León.
La región presenta más de 75000 hectáreas de viñedos con más de 5750
trabajadores. Tiene 530 bodegas que producen más de 200 millones de
litros de vino al año. La región presenta diez denominaciones de origen y
el 86% de los viñedos están dedicados a estos signos de calidad. En
2013, el valor económico de este sector alcanzó los 738 millones de
euros, siendo 134 generados por la exportación.
Los principales objetivos de las tecnologías llevadas a cabo durante el

proceso de obtención del vino están centrados en una extracción
efectiva de los compuestos fenólicos que son los responsables del color,
sabor y aromas de los vinos y el control de la fermentación del azúcar
que está directamente relacionado con el grado alcohólico del producto
final.
Como materia prima de la industria enológica la calidad de las uvas debe

ser controlada, ya que ésta influye directamente en la calidad del vino.
Para obtener uvas con las mejores características es necesario controlar,
no sólo la uva madura sino también el proceso de maduración de la
misma.
Tradicionalmente el control de la maduración de la uva se lleva a cabo

mediante la evaluación de una serie de correlaciones teóricas
denominadas “índices de maduración” consistentes en ciertas
formulaciones matemáticas propuestas para estimar el punto óptimo de
maduración de la uva.
261
Resumen en Español
Los índices de maduración externos se basan en el análisis organoléptico

de las bayas, determinando características propias de la maduración
como el color del grano y el peso del racimo mientras que, los índices de
maduración químicos se basan en la determinación analítica de los
compuestos más característicos que aumentan o disminuyen en el
proceso de maduración como la acidez o la concentración de azúcares.
El método de cata complementa los análisis físico-químicos que
tradicionalmente se llevan a cabo (contenido fenólico, grado, azúcares,
acidez, etc.) mediante métodos espectroscópicos, ópticos o
electroquímicos.
Existe una necesidad de desarrollar nuevos métodos objetivos y rápidos

que permitan establecer el momento óptimo de la maduración, un
factor fundamental para la calidad del vino.
En los últimos años se ha desarrollado un novedoso instrumento

llamado lengua electrónica que opera de manera análoga a los sentidos
humanos pudiendo percibir sabores y ser aplicada al análisis de líquidos
complejos. Según la definición dada por la IUPAC, estos sistemas se
basan en la combinación de sensores electroquímicos no selectivos con
gran selectividad cruzada acoplados con métodos quimiométricos (PCA,
LDA, ANN). El sistema es capaz de detectar simultáneamente una gran
cantidad de compuestos proporcionando información global de la
muestra, en lugar de información de componentes específicos
La mayoría de los trabajos en el campo de las lenguas electrónicas

aplicadas a la enología utilizan redes de sensores potenciométricos
donde se mide el potencial de membrana creado por la difusión de iones
a través de una membrana selectiva. También se han descrito sensores
impedimétricos y amperométricos modificados con diferentes
materiales como ftalocianinas, polímeros conductores, nanopartículas…
La posibilidad de utilizar electrodos voltamétricos es especialmente

interesante debido a su alta sensibilidad y versatilidad, ya que es posible
modular el rango de potencial, la forma de la curva y el material del
262
Resumen en Español
electrodo. Esta versatilidad permite obtener diferentes sensores con

diferente selectividad y sensitividad.
La selectividad de estos dispositivos se puede aumentar modificando la

superficie de los mismos con materiales electroactivos. Las interacciones
que ocurren entre el electrodo y la disolución pueden aumentar
extraordinariamente la selectividad del análisis. Esas interacciones
incluyen:
i) El carácter oxidante o reductor de la disolución puede

modificar el potencial de oxidación del material electródico.
ii) La actividad electrocatalítica del material del electrodo puede

facilitar la oxidación de los compuestos disueltos en la disolución de
estudio.
iii) La respuesta del material del electrodo está relacionada con

la capacidad del sensor para permitir la difusión de los iones entre la
disolución y la masa electrónica, este flujo es necesario para conservar la
electroneutralidad macroscópica del electrodo.
Pueden usarse diferentes familias de materiales como modificadores,

por ejemplo, polímeros conductores, ftalocianinas, nanopartículas… Se
ha demostrado que el empleo de diferentes materiales sensibles con
reactividad complementaria mejora la selectividad cruzada del conjunto
de sensores y aumenta su capacidad de discriminación (Figura 1).
Figura 1. Ilustración de la selectividad cruzada obtenida con sensores

voltamétricos modificados con materiales electroactivos diferentes. Red
expuesta a la misma sustancia.
263
Resumen en Español
Durante los últimos años, se han hecho muchos esfuerzos para

desarrollar biosensores electroquímicos enzimáticos para la detección
de diversos compuestos y en especial antioxidantes y azúcares. Las
enzimas reaccionan con sustratos específicos consumiendo oxígeno
(oxidasas) o produciendo la forma reducida del NAD(P)H
(deshidrogenasas). Estas transformaciones pueden medirse mediante
métodos electroquímicos.
Los sensores y biosensores pueden ser preparados mediante diferentes

técnicas (pasta de carbono, spin coating, screen printed…). Dependiendo
del método utilizado, se pueden obtener diferentes morfologías en la
superficie del electrodo que influyen directamente en el
comportamiento del mismo y en la respuesta obtenida.
El uso de nanotecnología aporta nuevas perspectivas en el desarrollo de

sensores ya que permite obtener dispositivos con una alta relación
superficie-volumen, facilitando la difusión de iones dentro de la película.
Las películas nanoestructuradas preparadas mediante self-assemblig
(SAM), Layer by Layer (LbL) o Langmuir-Blodgett (LB) han mostrado una
interesante capacidad como sensores para una gran variedad de
sustancias, con cinéticas más rápidas que los electrodos tradicionales.
Los éxitos en investigaciones previas en la aplicación de lenguas

electrónicas en el campo de la enología han hecho al grupo profundizar
en esta línea de investigación desarrollando nuevas redes de sensores
aplicadas en el control de calidad y madurez de la uva.
2- OBJETIVOS
De acuerdo con estas ideas generales los objetivos de la presente tesis
son:
1- Diseñar nuevos sensores y biosensores voltamétricos modificados

con diferentes materiales electrocatalíticos (nanopartículas,
ftalocianinas…) y enzimas (tirosinasa, lacasa, glucosa oxidasa y D-
fructosa deshidrogenasa).
264
Resumen en Español
2- Desarrollar nuevas tecnologías para depositar los materiales

sesibles y para inmobilizar los enzimas. Los métodos varían desde
técnicas simples que permiten preparar sensores sencillos y baratos
hasta técnicas más sofisticadas que producen sensores
nanostructurados con mejores prestaciones pero a un coste mayor.
3- Aplicar estos sensores en el análisis de diferentes analitos de

interés en la industria enológica como antioxidantes, azúcares y ácidos
orgánicos.
4- Combinar los sensores y biosensores para obtener lenguas

(bio)electrónicas aplicadas al análisis de mostos.
5- Desarrollar una lengua bioelectrónica que contenga enzimas para

la detección de fenoles y azúcares y aplicarla en el control de
maduración de uvas.
3- METODOLOGÍA Y RESULTADOS
Para alcanzar los objetivos fijados en esta tesis se han llevado a cabo los
siguientes experimentos, obteniendo los resultados que se presentan a
continuación:
3.1. Sensores nanostructurados mediante la técnica de

Langmuir-Blodgett
La técnica de Langmuir-Blodgett permite controlar la estructura del
sensor a nivel nanométrico aumentando la superficie de interacción
entre el electrodo y el analito. Empleando esta técnica se han
desarrollado cinco trabajos a lo largo de esta tesis.
El primer trabajo consistió en el uso de nanopartículas de oro

funcionalizadas con dodecanotiol como modificador. En primer lugar se
evaluó la respuesta de las nanopartículas frente a electrolitos básicos
como KCl, KBr, KNO3, KClO4 y MgCl2. Los voltamogramas obtenidos en
electrolitos que contenían potasio fueron muy parecidos y mostraban
los picos característicos de las nanopartículas sin embargo, en presencia
de cationes divalentes la intensidad de estos picos se vio reducida. De
estos resultados se pudo concluir que son los cationes quienes difunden
265
Resumen en Español
en el electrodo para mantener la electroneutralidad además, el estudio

del comportamiento dinámico indicó que el proceso está controlado por
transferencia electrónica. Cuando se analizaron ácidos orgánicos se
observaron los picos típicos de los protones a potenciales negativos
siendo posible distinguir entre ácidos monopróticos (tartárico) y
dipróticos o tripróticos (málico y cítrico). En el caso de los ácidos
fenólicos, los voltamogramas estaban dominados por los picos
correspondientes al proceso oxidativo de los fenoles en un rango de
potenciales entre 0.5-0.8 V. En ambos casos el proceso estuvo
controlado por la difusión de iones y los límites de detección
encontrados para todas las especies analizadas fueron de 10-5-10-6 mol·
L-1. En último lugar se llevó a cabo el análisis de mezclas de ácido
tartárico y ácido cafeico siendo posible la determinación de ambos sin
ninguna interferencia.
Figura 4. Voltamograma registrado empleando el sensor de AuNP-LB en mezclas

de cafeico:tartárico: 40:10 (línea verde), 25:25 (línea azul) y 10:40 (línea roja).
En segundo lugar se analizaron las propiedades electroquímicas de bis-

octacloro-ftalocianato de lutecio (LuPc2Cl32) y se compararon con
resultados obtenidos con sensores modificados con bisftalocianina de
lutecio (LuPc2). También se prepararon sensores mediante la técnica de
Langmuir-Schaefer comprobando si la deposición de las películas sobre
el electrodo influía en las propiedades finales del sensor. La
266
Resumen en Español
caracterización estructural se llevó a cabo mediante UV-Visible,

espectrometría de infrarrojo y de Raman. Los resultados obtenidos para
ambos tipos de películas fueron muy similares aunque difirieron de
aquellos encontrados en disolución o para películas no organizadas
realizadas mediante la técnica de casting. Las propiedades
electroquímicas de los sensores fueron testadas frente a una disolución
de catecol. La presencia de LuPc2Cl32 y LuPc2 provocó un desplazamiento
de los picos relacionados con los procesos redox del catecol hacia
potenciales menores y un aumento en la intensidad de corriente,
evidenciando el efecto electrocatalítico de ambas. Además, la presencia
de los sustituyentes clorados produjo una mayor desplazamiento y un
aumento de corriente mayor (25% versus 10% para LuPc2). El análisis del
comportamiento dinámico evidenció un proceso controlado por
transferencia electrónica en ambos casos. Los límites de detección
calculados para este tipo de sensores fueron de 10-5 mol·L-1.
En un tercer trabajo, se empleó la técnica de Langmuir-Blodgett para

modificar electrodos con dos moléculas electrocatalíticas como son las
nanopartículas de oro (AuNPs) y la bisftalocianina de lutecio (LuPc2). La
formación de películas delgadas de LuPc2 permitió la inclusión de AuNP
solubles en agua mediante la adsorción de las mismas en la película de
Langmuir. La formación de películas estables y homogéneas fue
comprobada empleando microscopía de ángulo de Brewster (BAM), UV-
Visible y espectrometría de Raman, así como microscopía electrónica de
barrido (SEM) y de transmisión (TEM). El efecto electrocatalítico de
ambos materiales se caracterizó mediante el análisis de hidroquinona.
La presencia de ambos electrocatalizadores se tradujo en un
desplazamiento hacia potenciales menores de los picos asociados a los
procesos redox del antioxidante así como un aumento de la intensidad
de corriente. Los límites de detección encontrados para este tipo de
sensores fueron de 10-7 mol·L-1 corroborando así la mejora alcanzada. El
comportamiento dinámico de los sensores estuvo controlado por el
proceso de difusión. Además, cuando se compararon los resultados
obtenidos con sensores de LuPc2 con y sin AuNPs, se encontró que la
267
Resumen en Español
velocidad del proceso de transferencia era dos veces más rápida con la
presencia de AuNPs.
Figura 5. Método de preparación de películas de LuPc 2:DODAB/AuNPs: DODAB,

LuPc2, AuNPs.
A la hora de preparar biosensores preservar la actividad del enzima es

un paso clave. La técnica de Langmui-Blodgett permite preparar
películas formadas por lípidos que simulan las membranas celulares en
las cuales los enzimas son absorbidos, favoreciendo así su actividad. En
un cuarto trabajo, se desarrolló una lengua bioelectrónica destinada al
análisis de fenoles y como objetivo final, la discriminación de mostos.
Para ello se empleó ácido araquídico como lípido para simular la
membrana celular, LuPc2 como mediador electrónico y tirosinasa y
lacasa como enzimas específicos para el análisis de antioxidantes. La
caracterización estructural de las películas y de los sensores se llevó a
cabo mediante isotermas, BAM y miscroscopía de fuerza atómica (AFM).
Todas las técnicas corroboraron la presencia del enzima tanto en las
películas como en los sensores. En primer lugar se realizó la
caracterización electroquímica de los electrodos frente a disoluciones de
distintos tipos de fenoles (mono, di y tri-fenoles). Todos los sensores
proporcionaron una respuesta diferente demostrando la selectividad
cruzada de la red y evidenciando la especificidad enzima-sustrato. Los
límites de detección alcanzados fueron del orden 10-7-10-8 mol·L-1. La
PCA de estos resultados demostró que la red no sólo es capaz de
discriminar las muestras sino que éstas aparecen organizadas en función
de los grupos fenólicos presentes en su estructura. Cuando la lengua
bioelectrónica fue empleada para el análisis de mostos, las muestras
fueron discriminadas en función de su contenido polifenólico.
268
Resumen en Español
Debido al buen funcionamiento de la red anterior, decidimos ampliarla

con enzimas específicos para el análisis de azúcares y llevar a cabo en
análisis de mostos desde el envero hasta post-vendimia. La
caracterización de las películas y sensores con glucosa oxidasa y d-
fructosa deshidrogenasa se realizó mediante isotermas y BAM. El
incremento en el área por molécula encontrado con la inclusión de los
enzimas en la película y el cambio en la rugosidad de las mismas
confirmó la absorción de las proteínas en la monocapa de Langmuir. En
un primer set de experimentos se analizaron mostos de 5 variedades de
uvas cosechados en dos años consecutivos (2012 y 2013). Los
voltamogramas mostraron picos asociados tanto a fenoles como a
azúcares obteniendo diferentes respuestas en función del sensor
empleado. La PCA de esto resultados mostró que la red era capaz no
solo de discriminar los mostos sino de agruparlos por variedades, a
pesar de las añadas, y en función de su composición química. Este
resultado evidenció que la lengua bioelectrónica produce una respuesta
muy semejante a los análisis tradicionales donde las características
fundamentales de cada variedad se mantienen a lo largo de las añadas.
Cuando se analizaron muestras con diferente grado de maduración se

observó que los cambios encontrados en los voltamogramas estaban de
acuerdo con los cambios en la composición química de la uva. Los
resultados de la PCA mostraron la capacidad de la red para seguir el
proceso de maduración de la baya. Los clusters pertenecientes a las
muestras analizadas desde el envero hasta la vendimia se encuentran
colocados en el sentido de las agujas del reloj, mientras que las muestras
cosechadas después de la vendimia no siguen esa tendencia. En último
lugar se llevaron a cabo correlaciones entre los resultados
electroquímicos obtenidos con la lengua bioelectrónica y los parámetros
químicos tradicionales encontrando excelentes correlaciones con los
parámetros relacionados con el azúcar, los fenoles e incluso el pH
debido a su influencia en el comportamiento de los enzimas.
269
Resumen en Español
Figura 6. Gráfico de Scores que ilustra la capacidad de discriminación de la

lengua bioelectrónica inmersa en mostos de la variedad Tempranillo (año
2013). 1-primera semana (envero), 2-segunda semana, 3- tercera semana, 4-
cuarta semana (vendimia) y 5- quinta semana (post-vendimia).
3.2. Sensores de pasta de carbono

Los sensores de pasta de carbono se prepararon mediante la mezcla de
un material carbonaceo con un aglutinante y añadiendo a la mezcla, en
caso necesario, un modificador. La pasta obtenida se comprimió en el
interior de un tubo de PVC empleando como contacto un hilo de cobre.
En esta tesis se han desarrollado dos trabajos diferentes basados en
electrodos de pasta de carbono.
En primer lugar se desarrollaron sensores formados por diferentes

materiales carbonaceos (polvo de grafito, microesferas de carbono y
nanotubos de carbono) incluyendo ftalocianina de dilitio (Li2Pc) como
modificador. Las respuestas voltamétricas obtenidas frente a una
disolución de ácido cítrico mostraron ser dependientes del material
carbonaceo empleado como matriz del sensor. Li2Pc presentó un efecto
electrocatalítico aumentando las intensidades de corriente de los picos
asociados al ácido cítrico en un orden de magnitud comparado con los
electrodos de carbono sin modificar. Además, la combinación de
nanotubos de carbono con Li2Pc produjo un efecto sinérgico que mejoró
270
Resumen en Español
el efecto electrocatalítico frente al ácido cítrico. El análisis del

comportamiento dinámico mostró que las respuestas de los tres
sensores estaban controladas por un mecanismo de difusión.
En segundo lugar se desarrolló una lengua electrónica formada por

sensores y biosensores modificados con distintas ftalocianinas cuyo
objetivo final fue la discriminación de variedades de uvas tintas. Los
sensores de pasta de carbono se modificaron con distintas ftalocianinas
metálicas: ftalocianina de cobre (CuPc), ftalocianina de cobalto (CoPc),
ftalocinina de cinc (ZnPc) and bisftalocianina de lutecio (LuPc2). Los
biosensores se prepararon mediante la técnica de casting and cross-
linking. Una gota de la disolución del enzima deseado (tirosinasa o
glucosa oxidasa) se depositó en la superficie del electrodo y una vez que
el disolvente se evaporó, se llevó a cabo el proceso de cross-linking con
una disolución de glutaraldehído. En un primer set de experimentos se
analizó la respuesta de los diferentes sensores frente a disoluciones
patrón de glucosa y catecol a dos pHs diferentes (3.1 y 7.0)
observándose que la respuesta obtenida dependió del metal central de
la ftalocianina y del pH de la disolución. La inclusión de los enzimas
demostró una mejora en la detección de los analitos, catecol en el caso
de la tirosinasa y glucosa para glucosa oxidasa, incrementando las
intensidades de los picos asociados con sus procesos redox y
desplazando los mismos hacia potenciales menores. Una vez
comprobado el buen funcionamiento de los sensores en la detección de
azúcares y fenoles, fueron empleados para el análisis de mostos. Para
disminuir la complejidad de la muestra, los mostos a analizar fueron
diluidos 1:2 en agua. Los voltamogramas mostraron picos relacionados
tanto con los fenoles como con los azúcares aunque debido a la
complejidad del medio estos fueron más anchos que los encontrados
para las disoluciones patrón. Cada electrodo mostró una respuesta
diferente para cada muestra analizada evidenciando la alta selectividad
cruzada alcanzada con la red. El Análisis de Componentes Principales
(PCA) demostró que la lengua electrónica era capaz de discriminar las
muestras de mostos. Este resultado mejoró cuando los biosensores
271
Resumen en Español
fueron incluidos en la red, organizando las muestras en el gráfico en

función de su composición química.
Figura 3. PCA de los resultados obtenidos del análisis de mostos con la lengua
bioelectrónica. a) Gráfico de Scores. 1- Tempranillo, 2- Prieto Picudo, 3- Mencía,
4- Cabernet-Sauvignon, 5- Garnacha. b) Gráfico de loadings donde aparecen 10
variables por cada sensor de la red.
3.3. Sensores serigrafiados

Electrodos serigrafiados desechables fueron modificados con enzimas
(tirosinasa y glucosa oxidasa) para formar una lengua bioelectronica
dedicada a la discriminación de mostos. Los mediadores electrónicos
empleados en el estudio fueron: carbono, platino, oro, grafeno, azul de
Prusia y nanopartículas de óxido de níquel. Los biosensores fueron
modificados con enzimas mediante la técnica de casting seguida de
cross-linking. Un gota de la disolución del enzima deseado (tirosinasa o
glucosa oxidasa) se depositó en la superficie del electrodo de trabajo y
una vez que el disolvente se evaporó se llevó a cabo el proceso de cross-
linking con vapores de glutaraldehído. Durante este proceso, el
electrodo de referencia y el contraelectrodo fueron protegidos por una
máscara para evitar su contaminación. Como experimento preliminar,
los electrodos fueron empleados en el análisis de disoluciones patrón de
fenoles y azúcares (catecol y glucosa). Todos los sensores mostraron los
picos característicos de los analitos produciéndose un aumento en la
intensidad y un desplazamiento de éstos hacia potenciales menores
cuando fueron analizados con los biosensores. Se observaron grandes
272
Resumen en Español
diferencias en las respuestas en función del mediador electrónico

empleado siendo grafeno, nanopartículas de óxido de niquel y azul de
Prusia los modificadores que presentaron mejor comportamiento y
mayor efecto electrocatalítico.
Cuando la lengua (bio)electrónica fue empleada para analizar mostos de

diferentes variedades se obtuvieron voltamogramas con diversos picos:
unos de ellos relacionados con las especies electroactivas presentes en
la muestra (antioxidantes y azúcares) y otros relacionados con el
modificador electrónico (ej. azul de prusia). Los mostos fueron diluidos
1:2 en tampón fosfato para disminuir la irreproducibilidad de la medida
debido a la complejidad de la muestra. La interacción entre los
electrodos y el mosto produjo respuestas muy complejas y diferentes
para cada uno de ellos, demostrando el alto grado de selectividad
cruzada alcanzado con la red de electrodos. La capacidad de
discriminación de la red de sensores fue analizada mediante PCA. Cinco
cálculos diferentes fueron llevados a cabo: i) red de sensores sin
modificar, ii) red de sensores de tirosinasa, iii) red de sensores de
glucosa oxidasa, iv) ambas redes de biosensores juntas y v) todas las
redes juntas. Las redes por separaron mostraron una discriminación
parcial de las muestras que se vio mejorada en las redes de biosensores,
sin embargo, cuando las respuestas de las dos redes de biosensores son
tratadas conjuntamente las muestras no sólo son discriminadas sino que
en el gráfico aparecen ordenadas en función de su composición química.
Este sistema mostró una discriminación muy próxima a la alcanzada por
otros sistemas mucho más caros y complejos. En último lugar, se
correlacionaron los resultados obtenidos con los parámetros químicos
tradicionales encontrando muy buenas correlaciones con el grado Brix y
el Índice Total de Polifenoles.
3.4. Sensores modificados con nanopartículas core-shell

Nanopartículas bimetálicas de Au-In con diferentes composiciones se
sintetizaron mediante sputtering magnético de cantidades controladas
de Au y de In sobre electrodos de ITO. El análisis de la morfología y la
composición de las nanopartículas se realizó mediante microscopía
273
Resumen en Español
electrónica de transmisión (TEM), microscopía electrónica de barrido

(SEM), espectrometría de rayos X (XPS) y espectrometría de dispersión
de energía de rayos X (EDS). Las nanopartículas presentaron centros de
Au-In y mezclas de Au-Óxido de In en la capa exterior. La voltametría
cíclica se empleó para la caracterización electroquímica de los diferentes
sensores frente a una disolución de hidroquinona. La presencia de
nanopartículas Au-In produjo un desplazamiento de los picos asociados
a los procesos redox de la hidroquinona demostrando así su efecto
electrocatalítico. Las respuestas voltamétricas obtenidas con cada uno
de los sensores dependieron de la proporción Au-In presente en las
nanopartículas. Mientras que las nanopartículas con proporciones Au:In
1:1 y 1:2 dieron lugar a un proceso redox de dos electrones, cuando la
proporción fue 3:1 el mecanismo del proceso redox encontrado fue de
dos pasos consecutivos de un electrón. El análisis del comportamiento
dinámico de los electrodos de nanopartículas mostró que la reacción
electroquímica está controlada por el mecanismo de difusión, indicando
que la alta relación superficie volumen de las nanopartículas facilitó una
rápida transferencia electrónica entre la superficie del electrodo y el
analito. Los límites de detección encontrados para estos sensores
estuvieron en el rango de 10-5-10-6 mol·L-1.
274
Resumen en Español
Figura 2. Voltametría cíclica de a )ITO, b)electrodo de Au, c)nanopartículas de

Au, d) nanopartículas de AuIn2, e)nanopartículas de AuIn, y f)nanopartículas de
-3 -1
Au3In inmersas en una disolución de hidroquinona 10 mol·L .
4- CONCLUSIONES
De acuerdo con los objetivos establecidos, las conclusiones obtenidas en
el desarrollo de esta tesis son:
1- Se han diseñado y construido un conjunto de sensores y

biosensores voltamétricos mediante diferentes técnicas y empleando
diferentes materiales electrocatalíticos y se han aplicado en el análisis
de diferentes analitos presentes en mostos como antioxidantes, ácidos
orgánicos y azúcares
a) Nanopartículas de oro funcionalizadas con n-dodecanotiol se

emplearon para modificar electrodos mediante la técnica de Langmuir-
275
Resumen en Español
Blodgett. Los electrodos mostraron un gran efecto electrocatalítico

frente a ácidos fenólicos y orgánicos siendo capaces de proporcionar
información sobre mezclas de ácido cafeico/tartárico sin ninguna
interferencia.
b) Las técnicas de Langmuir-Blodgett y Langmuir-Schaefer se

emplearon para el desarrollo de sensores con bis-octacloro-ftalocianato
de lutecio. Las películas presentaron buena estabilidad y sensibilidad
alcanzando límites de detección de 10-5 mol·L-1 para la detección de
catecol.
c) Dos materiales electrocatalíticos fueron empleados para

modificar un electrodo de óxido de indio y estaño (ITO) mediante la
técnica de Langmuir-Blodgett. Se observó un efecto sinérgico
electrocatalítico en el análisis de hidroquinona, obteniendo un aumento
de la corriente y alcanzando límites de detección del orden de 10-7 mol·L-
1
.
d) La influencia de la estructura del material cabonaceo

empleado en el desarrollo de sensores de pasta de carbono ha sido
evaluada en el análisis de ácido cítrico. La respuesta obtenida fue
dependiente del material empleado, encontrando un efecto sinérgico
entre la ftalocianina de litio empleada como modificador y los
nanotubos de carbono.
e) Nanopartículas core-shell de Au-In fueron preparadas

mediante sputtering magnético. Estos sensores se emplearon en la
detección de hidroquinona observándose un gran efecto
electrocatalítico y alcanzando límites de detección del orden de 10-5-10-6
mol·L-1. La actividad electrocatalítica y el mecanismo del proceso redox
dependieron de la cantidad de oro estabilizada en la superficie.
2- Las redes de sensores y biosensores se emplearon en el análisis

de mostos. Para conocer el funcionamiento de los electrodos fueron
empleados previamente para el análisis de disoluciones patrón. Tres
lenguas (bio)electrónicas se desarrollaron con éste propósito.
276
Resumen en Español
a) Una lengua (bio)electrónica formada por sensores y

biosensores de pasta de carbono modificados con diferentes
metaloftalocianinas y enzimas. La respuesta obtenida dependió del
metal central de la ftalocianina y mejoró con la inclusión de enzimas.
b) Una red formada por electrodos comerciales miniaturizados

con diferentes mediadores electrónicos modificados con enzimas. Los
resultados obtenidos presentaron buenas correlaciones con el Brix y el
Índice Total de Polifenoles. Las ventajas de esta legua electrónica son el
bajo precio de los sensores, la facilidad de uso y la portabilidad.
c) Un lengua bioelectrónica nanoestruturada formada por

sensores modificados con fenol oxidasas empleando bisftalocianina de
lutecio como mediador electrónico, desarrollada mediante la técnica de
Langmuir-Blodgett. Los límites de detección obtenidos en el análisis de
disoluciones patrón de fenoles fueron de 10-8 mol·L-1.
Las redes fueron capaces de discriminar las muestras de uva en función

de su contenido polifenólico y de la concentración de azúcares.
3- Una lengua bioelectrónica formada por sensores con enzimas

para el análisis de fenoles y azúcares fue empleada para controlar el
proceso de maduración de la uva. La técnica empleada para desarrollar
los sensores fue Langmuir-Blodgett favoreciendo el funcionamiento del
enzima mediante la formación de un entorno biomimético. Las muestras
fueron cosechadas semanalmente desde el envero hasta post-vendimia.
Los resultados electroquímicos obtenidos fueron correlacionados con los
parámetros químicos analizados por los métodos tradicionales mediante
PLS-1 obteniendo excelentes correlaciones para los parámetros
relacionados con fenoles y azúcares.
277
ANEX
Anex
The work carried out during this thesis period has led to the following
publications:
18- C. Medina-Plaza, J.A de Saja, J.A. Fernández-Escudero, E. Barajas, G.

Medrano, M.L. Rodriguez-Mendez. Array of biosensors for
discrimination of grapes according to grape variety, vintage and
ripeness. Analytica Chimica Acta. 2016, Submitted.
17- M.L. Rodríguez-Méndez, J.A. de Saja, C. Medina-Plaza, C. Garcia-

Hernandez. Electrochemical Sensors for the Detection of Antioxidants. In
Bioactive compounds: natural sources, physicochemical characterization,
applications. Eds. C. Apetrei. Ed. Benthan Science Publishers. Sharjah.
United Arab Emirates. 2016, Accepted.
16- M.L. Rodriguez-Mendez, C. García-Hernandez, C. Medina-Plaza, C.

García-Cabezón, J.A. de Saja. Application of multisensor systems based
on phthalocyanines in oenology. Case study: monitoring the quality and
ripeness of grapes. Journal of Porphyrins Phthalocyanines. 2016,
Accepted.
15- M.L. Rodríguez-Méndez, J.A. de Saja, C. Medina-Plaza, C. Garcia-

Hernandez. Electronic tongues for the organoleptic characterization of
wines. In Electronic tongues and noses in food science. Eds. M.L.
Rodriguez-Mendez. Ed. Elsevier, San Diego. 2016, 265-273.
14- M.L. Rodriguez-Mendez, C. Medina-Plaza, C. García-Hernández, S.

Rodriguez, C. García-Cabezon, D. Paniagua, M.A. Rodriguez-Perez, J.A de
Saja. Improvement of electrocatalytic effect in voltammetric sensors
based on phthalocyanines. Journal of Porphyrins Phthalocyanines. 2016,
20, 1-8.
13- C. Medina-Plaza, M.L Rodriguez-Mendez, P. Sutter, X. Tong, E.

Sutter. Nanoscale Au-In alloy-oxide core-shell particles as
electrocatalysts for efficient hydroquinone detection. Journal of Physical
Chemistry-C. 2015, 119 (44), 25100–25107.
281
Anex
12- C. Garcia-Hernandez, C. Medina-Plaza, C. Garcia-Cabezon, F. Martin-

Pedrosa, I. del Valle, J.A. de Saja, M.L. Rodriguez-Mendez. An
electrochemical quartz crystal microbalance multisensor system based
on phthalocyanine nanostructured films: Discrimination of musts.
Sensors. 2015, 15, 29233-29249.
11- C. Garcia-Hernandez, C. Garcia-Cabezon, C. Medina-Plaza, F. Martin-

Pedrosa,Y. Blanco, J. A. deSaja, M.L Rodriguez-Mendez. Electrochemical
behavior of polypyrrol/AuNP composites deposited by different
electrochemical methods. Sensing properties towards catechol. Beilstein
Journal of Nanotechnology. 2015, 6, 2052-2061.
10- C. Medina-Plaza , C. Garcia-Hernandez, J.A. de Saja, J.A. Fernandez-

Escudero, E. Barajas-Tola, G. Medrano, C. García-Cabezón, F. Martín-
Pedrosa, M.L. Rodriguez-Mendez. The advantages of disposable screen-
printed biosensors in a bioelectronic tongue for the analysis of grapes.
LWT- Food Science and Technology. 2015, 62, 940-947.
9- M.L. Rodriguez-Mendez, C. Medina-Plaza, J.A. de Saja, J.A. Fernandez-

Escudero, E. Barajas-Tola, G. Medrano. Analysis of grapes and wines
using a voltammetric bioelectronics tongue. Correlation with the
phenolic and sugar content. IEEE Sensors Journal. 2014, IEEE Conference
Publications, 2139-2142.
8- C. Medina-Plaza, C. García-Cabezón, C. García-Hernandez, C.

Bramorski, Y. Blanco-Val, F. Martín-Pedrosa, T. Kawai, J.A. de Saja, M.L.
Rodriguez-Mendez. Analysis of organic acids and phenols of interest in
the wine industry using Langmuir-Blodgett films based on functionalized
nanoparticles. Analytica Chimica Acta. 2014, 853, 572-578.
7- C. Medina-Plaza, L.N. Furini, C.J.L. Constantino, J.A. de Saja, M.L.

Rodriguez-Mendez. Synergistic electrocatalytic effect of nanostructured
mixed films formed by functionalized gold nanoparticles and
bisphthalocyanines. Analytica Chimica Acta.2014, 851, 85-102.
6- C. Medina-Plaza, M.L. Rodriguez-Mendez, J.A. de Saja. Bioelectronic

tongue based on lipidic nanostructured layers containing phenol
282
Anex
oxidases and lutetium bisphthalocyanine for the analysis of grapes.

Biosensors&Bioelectronics. 2014, 57, 276-283.
5- M.L. Rodríguez-Méndez, C. Apetrei, M. Gay, C. Medina-Plaza, J.A. de

Saja, S.Vidal, O.Aagaard, M.Ugliano,J. Wirth,V. Cheynier. Evaluation of
oxygen exposure levels and polyphenolic content of red wines using an
electronic panel formed by an electronic nose and an electronic tongue.
Food Chemistry. 2014, 155, 91-97.
4- P.Alessio, C.Apetrei, R.J.G.Rubira, C.J.L.Constantino, C.Medina-Plaza,

J.A.de Saja, M.L. Rodríguez-Méndez. Structural and electrochemical
properties of a bis-octachloro-phthalocyaninate nanostructured films.
Application as voltammetric sensor. J. Nanoscience and
Nanotechnology.2014, 14, 6574-6773.
3- M.L. Rodríguez-Méndez, C. Apetrei, C. Medina-Plaza, R. Muñoz, J.A.

de Saja. Sensor array based on phthalocyanines: New developments on
nanostructured and biomimetic electrochemical sensors. In Multisensor
systems for chemical analysis: materials and sensors. Eds. L. Lvova, D.
Kirsanov, C. Di Natale, and A. Legin. Ed. Pan Standford Publishing Pte.
Ltd. Singapore, 2014, 139-179.
2- C. Medina-Plaza, G.Revilla, R.Muñoz, J.A. Fernandez-Escudero,

E.Barajas, G.Medrano, J.A. de Saja, M.L. Rodríguez-Méndez. Electronic
tongue formed by sensors and biosensors containing phthalocyanines as
electron mediators. Application to the analysis of red grapes. Journal of
Porphyrins and Phthalocyanines. 2013, 17, 1-11.
1- C. Apetrei, C. Medina-Plaza, M.L. Rodríguez-Méndez, J.A. de Saja.

Electrochemical characterization of dilithium phthalocyanine
carbonaceous electrodes. Journal of Porphyrins and
Phthalocyanines.2013, 17, 1-7.
283

Tesis Cristina Medina

Cargado por

Copyright:

Formatos disponibles

Tesis Cristina Medina

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Tesis Cristina Medina

Cargado por

Copyright:

Formatos disponibles

Sensorarrays

DEPARTAMENTO DE QUÍMICA FÍSICA Y QUÍMICA

Sensor arrays for Enology applications:

Presentada por Cristina Medina Plaza para optar al

Prof. Dra. María Luz Rodríguez Méndez

Dª María Luz Rodríguez Méndez, con D.N.I: nº 50305983-S,

Valladolid, 15 de Abril de 2016

Fdo. María Luz Rodríguez Méndez

List of Abbreviations and Acronyms……………………………………………………1-4

Justifications and Objectives…………………………………………………………5-10

1.1.1 Chemical composition of the grapes………………….…………..15

1.1.2 Berry ripening……………………………………………………..………….16

1.2 ELECTRONIC TONGUE………………………………………………………..19

1.2.1- Mass sensors……………………………………………………..………….20

1.2.2- Optical sensors……………………………………………..……………….21

1.2.5-Preparation of voltammetric sensors..……………………………32

CHAPTER 2- Analysis of Organic Acids and Phenols of Interest in the

2.2 MATERIALS AND METHODS……………………………………………….49

2.2.5- Electrochemical studies…………………………………………………51

2.3 RESULTS AND DISCUSSION…………………………………………………51

2.3.2-Electrochemical response in simple electrolytes…………….52

2.3.3-Electrocatalytic effect of SDODAuNP-LB towards organic

2.3.4-Electrocatalytic effect of SDODAuNP-LB towards phenolic

2.3.5-Response of SDODAuNP-LB towards mixtures of organic

2.3.6-Electrode repeatability and reproducibility…………..………..62

CHAPTER 3- Structural and Electrochemical Properties of Lutetium Bis-

3.2 MATERIALS AND METHODS……………………………………………….73

3.2.2-Langmuir, LB and LS films…………………..….……………………….73

3.3.2-LB and LS films. Growth monitored by UV-Vis

3.3.3-LB and LS films. Molecular organization determined by

3.3.4-Surface Enhancement Resonance Scattering

3.3.6-Electrochemical detection of catechol……………………………83

CHAPTER 4- Synergistic Electrocatalytic Effect of Nanostructured Mixed

4.2 MATERIALS AND METHODS……………………………………..………..99

4.2.1-Langmuir monolayers and Langmuir-Blodgett films…..…..99

4.3 RESULTS AND DISCUSSION………………………………………………102

4.3.4-Electrocatalytic properties. Voltammetric sensors……….109

CHAPTER 5- Bioelectronic Tongue Based on Lipidic Nanostructured

5.2 MATERIALS AND METHODS……………………………………………..124

5.2.2-Langmuir and Langmuir-Blodgett films

5.2.4-Phenols and grapes……………………..…………………………..…..125

5.2.5-Statistical analysis. Data treatment………………..…………….126

5.3 RESULTS AND DISCUSSION………………………………………………126

5.3.1-Langmuir monolayers and Langmuir-Blodgett

5.3.2-Electrochemical response towards phenols………………....129

5.3.3-Array of sensors. Discrimination of phenolic

5.3.4-Array of sensors. Response towards grapes………............136

CHAPTER 6-Array of Biosensors for Discrimination of Grapes According

6.2 MATERIALS AND METHODS……………………………………………..147

6.3 RESULTS AND DISCUSSION………………………………………………150

6.3.1-Characterization of Langmuir-Blodgett films…………………150

6.3.2-Discrimination of must obtained from different varieties of

6.3.3-Monitoring the grape ripeness by means of the

6.3.4-Multiparametric correlations between the bioelectronics

7.2 MATERIALS AND METHODS……………………………………………..167

7.3 RESULTS AND DISCUSSION………………………………………………168

7.3.2-Analysis of simple electrolytes……………………………………..172

7.3.3-Analysis of citric acid…………………………………………………….174

CHAPTER 8-Electronic Tongue Formed by Sensors and Biosensors

8.2 MATERIALS AND METHODS……………………………………………..185