Development of Electrochemical Sensor

and Biosensor Platforms: Detection of

Therapeutic Drugs and Heavy Metal Ions

Sereilakhena Phal

Department of Chemistry

Doctoral Thesis

Umeå 2019

This work is protected by the Swedish Copyright Legislation (Act 1960:729) Dissertation for PhD ISBN: 978-91-7855-149-1 Cover picture: Sereilakhena Phal Electronic version available at: http://umu.diva-portal.org/ Printed by: Lars Åberg, KBC Service Center, Umeå University Umeå, Sweden 2019

Dedicated to my parents, sisters, and brothers

i

Table of Contents Table of Contents ........................................................................... i

Abstract ....................................................................................... iv

Enkel sammanfattning på svenska ............................................... vi

List of Abbreviations .................................................................. viii

List of Publications ........................................................................ x

1. Introduction .............................................................................. 1

1.1. Aims of the Thesis .................................................................. 3

1.2. Thesis Outline ........................................................................ 3

2. Background ............................................................................... 5

2.1. Sensors and Biosensors ........................................................... 5

2.1.1. Electrochemical Sensors and Biosensors .............................. 6

2.1.1.1. Electrochemical Immittance Spectroscopy ...................... 6

2.1.1.2. Stripping Voltammetry ................................................ 11

2.1.1.3. Cyclic Voltammetry .................................................... 13

2.1.2. Electrode Surface Modification ........................................... 14

2.2. Methotrexate (MTX) ............................................................... 16

2.3. Analytical Methods for MTX Detection ....................................... 17

2.4. Heavy Metal Ions (Pb2+ and Cd2+) ............................................ 21

2.5. Methods of Detection for Pb2+ and Cd2+ .................................... 21

3. Methods .................................................................................. 27

3.1. Electrochemical Setup ............................................................ 27

3.2. Electrode Cleaning ................................................................. 28

3.3. 4-Carboxybenzenediazonium Synthesis .................................... 28

3.4. Electrochemical Biosensor for Detection of MTX (paper I and II) .. 28

ii

3.4.1 Electrode Surface Modification ............................................ 28

3.4.2. Characterization .............................................................. 31

3.4.2.1. Electrochemical Immittance Spectroscopy ..................... 31

3.4.2.2. X-ray Photoelectron Spectroscopy (paper I) .................. 31

3.4.2.3. Cyclic Voltammetry (paper II) ..................................... 32

3.4.2.4. Infrared Spectroscopy (paper II) .................................. 32

3.4.3. Detection of MTX using EIS ............................................... 32

3.4.4. Multivariate Data Analysis ................................................. 33

3.5. Electrografting of 4-CBD on GCE (paper III) .............................. 34

3.5.1. Electrochemical Behaviour of 4-CBD on GCE ....................... 34

3.5.2. Electrochemical Oxidative Cleaning of Grafted Surface .......... 34

3.5.3. Effect of 4-CBD Concentration and Scan Rate on Grafting ..... 35

3.6. Electrochemical Sensor for Detection of Metal Ions (paper IV) ..... 35

3.6.1. Preparation of Carboxyphenly-GCE (CP/GCE) ...................... 35

3.6.2. Detection of Pb2+ and Cd2+ ................................................ 36

3.6.3. Detection of Pb2+ and Cd2+ in Tap Water ............................. 36

4. Results and Discussion ............................................................ 37

4.1. Paper I ................................................................................. 37

4.1.1. Characterization of Electrode Modification steps ................... 37

4.1.2. Detection of MTX using EIS ............................................... 37

4.2. Paper II ................................................................................ 38

4.2.1. Characterization of Electrode Modification Steps .................. 38

4.2.2. Detection of MTX using EIS of Redox Probe, Fe(CN)63-/4- ....... 38

4.2.3. Detection of MTX using EIS without Redox Probe ................. 39

4.3. Paper III ............................................................................... 41

iii

4.3.1. Electrochemical Behaviour of 4-CBD ................................... 41

4.3.2. Electrochemical Oxidative Cleaning of the Grafted Surface .... 42

4.3.3. Effect of 4-CBD Concentration and Scan Rate on Grafting ..... 43

4.4. Paper IV ............................................................................... 44

4.4.1. Procedure for Measurement of Cd2+ and Pb2+ and the

Performance of Different Electrodes ............................................. 44

4.4.2. Optimization of Relevant Parameters .................................. 47

4.4.3. Analytical Performance ..................................................... 47

4.4.4. Determination of Cd2+ and Pb2+ in Tap Water ...................... 48

5. Conclusions ............................................................................. 49

6. Future Perspectives ................................................................. 50

Acknowledgements ..................................................................... 51

References .................................................................................. 53

iv

Abstract

Electrochemical sensors and biosensors combine the sensitivity of

electroanalytical methods with the selectivity of a sensor or biosensor surface.

The chemical or biochemical component (receptor) in the sensor recognizes

an analyte and produces an electrical signal which is proportional to the

analyte concentration. Some of these sensors are routinely used in clinical

applications and are known for their simplicity, portability, cost-effective, and

miniaturization. The glucose sensor used in the management of diabetes is a

good example of such biosensors.

This thesis deals with the development of electrochemical biosensor and

sensor platforms for the detection of therapeutic drugs, demonstrated using

methotrexate (MTX) which is the most common drug used for the treatment

of cancer patients, and heavy metal ions (Pb2+ and Cd2+).

The biosensor surfaces were generated by immobilization of antibody (anti-

MTX) on chemically modified gold electrodes using different surface

modification protocols. Self-assemble monolayer (SAM) using alkanethiol

(cysteamine) or electrografting with diazonium salt (4-

carboxybenzenediaonium tetrafluoroborate, 4-CBD) was used for surface

modification. The surface modification was monitored and characterized

using electrochemical immittance spectroscopy (EIS) and cyclic voltammetry

(CV) along with other complementary technique such as X-ray photoelectron

spectroscopy (XPS). The biosensing surfaces were used for the detection of

MTX in an electrochemical flow cell (paper I) and in a batch system (paper II).

The detection was based on non-faradaic electrochemical immittance

spectroscopy (EIS) and singular value decomposition (SVD) for data

evaluation. Both electrochemical biosensors provided the lowest limit of

detection, LOD (at picomolar level) compared to earlier reports.

The electrografting of 4-CBD on glassy carbon electrode (GCE) using CV and

the parameters that influence the number of monolayers that can be grafted

on the surface are demonstrated (paper III). The CVs obtained during grafting

showed one or two reduction peaks, and this was found to be related to the

v

number of monolayers deposited on the electrode. One can increase the

number of monolayers by increasing the concentration of 4-CBD or

decreasing the scan rate. The GCE, grafted using 4-CBD, was incorporated

with Bi by an in situ electrodeposition of Bi3+ and used as an electrochemical

sensor for detection of Pb2+ and Cd2+ using square wave anodic stripping

voltammetry, SWASV (paper IV). The sensor resulted in LOD of 10 μg L-1 for

Pb2+ and 25 μg L-1 for Cd2+. The applicability of the sensor was tested for

detection of Pb2+ and Cd2+ in tap water and compared with ICP-OES. The

results were comparable, demonstrating the potential of the sensor as an

alternative to ICP-OES for the detection of metal ions in water samples.

vi

Enkel sammanfattning på svenska

Elektrokemiska sensorer är analytiska verktyg som används för att detektera

och kvantifiera kemiska ämnen. Vissa av dessa sensorer används rutinmässigt

i kliniska applikationer och är kända för bland annat sina enkelhet, snabbhet,

miniatyrisering och portabla format. Glukossensorn som används vid

hantering av diabetes är ett bra exempel på sådana biosensorer.

Denna avhandling handlar om utveckling av elektrokemiska biosensorer och

sensorplattformar för detektion av läkemedel som används inom kemoterapi

för cancerbehandling och för bestämning av metalljoner i vattenprover.

Kemoterapi är en cancerbehandlingsform med ett smalt terapeutiskt fönster,

det vill säga att dosen för en positiv anticancereffekt är mycket nära den som

orsakar allvarliga, ibland dödliga, biverkningar. Det är därför av största vikt

att dosen individanpassas så korrekt och precist som möjligt, baserat på den

farmakokinetiska profilen som erhålls. För att göra det mäts

läkemedelshalterna i kroppsvätskor, såsom plasma, serum eller blod.

De mätmetoder som används idag för analys av cytostatika och deras

metaboliter har många nackdelar, såsom långa analysprocedurer,

förbrukning av stora mängder organiska lösningsmedel och dyr

instrumentering.

Fokus i denna avhandling har varit att utveckla biosensorytor för detektion av

läkemedel som används för cancerbehandling. Mtotrexat (MTX) som är ett

av de vanligaste läkemedlen inom kemoterapi har använts som

modellsubstans. Konceptet har även utvidgats för utveckling av sensorytor för

detektion av metalljoner i vattenprover.

Biosensorytorna har framställts genom immobilisering av antikroppar (anti-

MTX) på kemiskt modifierade guldelektroder där olika

ytmodifieringsprotokoll har använts., Bioensorytorna har karakteriserats med

hjälp av elektrokemisk immittansspektroskopi (EIS) och cyklisk voltammetri

(CV) och röntgenfotoelektronspektroskopi (XPS). Biosensorerna har sedan

använts för detektion av MTX in en flödecell (artikel I) och i ett batchsystem

vii

(artikel II). Mättekniken är baserad på elektrokemisk immittansspektroskopi

(EIS) och databehandling enligt metoden ”singular value decomposition”

(SVD). De låga detektionsgränser för MTX (5×10-12 M) som erhållits är lägre

än vad som rapporterats tidigare.

Sensorytbehandling med ett diazoniumsalt (4-karboxybensendiazoniumsalt,

4-CBD) genom så kallad elektrograftning med hjälp av cyklisk voltametri (CV)

har utförts. Studien visar hur bindningen av 4-CBD på en elektrod av vitrös

grafit (glassy carbon electrode, GCE) påverkas av CV, vilket i sin tur påverkar

antalet monolager som kan läggas på elektrodytan (artikel III).

Elektrograftad GCE med insitu-deponering av Bi3+ har också använts som

elektrokemisk sensor för bestämning av Pb2+ och Cd2+ med hjälp av ”square

wave anodic stripping voltametry” SWASV (artikel IV). Sensorns

användbarhet testades för detektion av bly och kadmium i kranvatten och

jämfördes med induktivt kopplad plasma optisk emissionsspektrometri (ICP-

OES). Resultaten var jämförbara vilket visade sensorns potential som ett

alternativ till ICP-OES (som är mycket dyrare) för detektion av metalljoner.

viii

List of Abbreviations

AAS Atomic Absorption Spectroscopy

Ab Antibody

AC Alternating Current

ACN Acetonitrile

AdSV Adsorptive Stripping Voltammetry

AFM Atomic Force Microscope

AOAC Association of Official Analytical Chemists

ASV Anodic Stripping Voltammetry

BE Binding Energy

Bp Blocking Polymer

4-CBD 4-Carboxybenzenediazonium Tetrafluoroborate

CP Carboxyphenyl

CPE Constant Phase Element

CSV Cathodic Stripping Voltammetry

CV Cyclic Voltammetry

CVs Cyclic Voltammograms

Cys Cysteamine

DC Direct Current

DNA Deoxyribonucleic Acid

DP Differential Pulse

DPV Differential Pulse Voltammetry

ECD Electrochemical Detector

EDC N-(3-dimethylaminopropyl)-N′-ethylcarbodimide

EIS Electrochemical Immittance Spectroscopy/

Electrochemical Impedance Spectroscopy

EIS-MS Electrospray Ionization Mass Spectrometry

EIS-MS/MS Electrospray Ionization Tandem Mass Spectrometry

FAAS Flame Atomic Absorption Spectroscopy

GCE Glassy Carbon Electrode

Glu Glutaraldehyde

ix

HPLC High Performance Liquid Chromatography

ICP-MS Inductively Coupled Plasma Mass Spectrometry

ICP-OES Inductively Coupled Plasma Optical Emission Spectrometry

IR Infrared

LOD Limit of Detection

LS Linear Sweep

LSV Linear Sweep Voltammetry

MS Mass Spectrometry

MS/MS Tandem Mass Spectrometry

MTX Methotrexate

MVDA Multivariate Data Analysis

NHS N-Hydroxysuccinimide

PhB Phosphate Buffer Solution

RMSEC Root Mean Square Error of Calibration

RMSEP Root Mean Square Error of Prediction

RSD Relative Standard Deviation

SAMs Self-Assemble Monolayers

SPCE Screen Printed Carbon Electrode

SVD Singular Value Decomposition

SW Square Wave

SWASV Square Wave Anodic Stripping Voltammetry

SWV Square Wave Voltammetry

UV Ultraviolet

UV-Vis Ultraviolet-Visible

WHO World Health Organization

XPS X-ray Photoelectron Spectroscopy

x

List of Publications

I. Sereilakhena Phal, Britta Lindholm-Sethson, Paul Geladi, Adrey

Shchukarev, and Solomon Tesfalidet. Determination of methotrexate in

spiked human blood serum using multi-frequency electrochemical

immittance spectroscopy and multivariate data analysis. Analytica

Chimica Acta 987 (2017) 15-24.

II. Sereilakhena Phal, Besart Shatri, Avni Berisha, Paul Geladi, Britta

Lindholm-Sethson, and Solomon Tesfalidet. Covalently electrografted

carboxyphenyl layers onto gold surface serving as a platform for the

construction of an immunosensor for detection of methotrexate.

Electroanalytical Chemistry 812 (2018) 235-243.

III. Sereilakhena Phal, Kenichi Shimizu, Solomon Tesfalidet. New insight

into the electrochemical behavior of 4-carboxybenzenediazonium

tetrafluoroborate on glassy carbon electrode. (Submitted manuscript)

IV. Sereilakhena Phala, Huyền Nguyễna, and Solomon Tesfalidet. In situ

Bi/carboxyphenyl-modified glassy carbon electrode (Bi/CP/GCE) as a

sensor platform in square wave anodic stripping voltammetry: detection of

Pb2+ and Cd2+ in tap water. (Manuscript)

a: Joint first authors

1

1. Introduction

In recent years sensors and biosensors have gained more popularity in their

role in detecting and monitoring different types of analytes in the fields of

clinical analysis, medical diagnosis, environmental monitoring, and food

safety [1-4]. They are popular because they are associated with high

sensitivity, short analysis time, and lower cost compared to conventional

methods such as immunoassays, spectroscopy, and chromatography

techniques [1, 5]. Moreover they can be used by the general public to analyze

complex samples without tedious sample preparation. Powerful tools such as

the glucose biosensor, pregnancy test kits, cholesterol test kits, and some

blood analyzers are good examples of such sensors [6].

There are three kinds of sensors and biosensors which are classified according

to their transduction techniques: electrochemical, piezoelectric or optical [1].

Among these, sensors based on electrochemical techniques are known to be

easy to use, portable, and cost-effective [7]. Electrochemical immittance

spectroscopy (EIS) is the technique of choice in the development of biosensors

because it is sensitive and non-destructive for the sensor surface [7-10].

Stripping voltammetry has been widely utilized for detection of trace levels of

various heavy metals and compounds in environmental and clinical samples

as well as food analysis [11, 12]. The broad application of this method is due to

its sensitivity, selectivity, high accuracy, precision and less need of sample pre-

treatment.

The EIS can be based on faradaic and nonfaradaic processes. In faradaic EIS,

a redox probe such as Fe(CN)63-/4- is used to investigate its interaction on the

electrode surface. For nonfaradaic EIS, the measurement is made directly on

the solution containing the target compound without the redox probe [7, 13].

Since no redox probe is needed, development of biosensors based on

nonfaradaic EIS is a perfect choice for direct measurement on samples such

as at a point-of-care [7, 14, 15]. The detection is based on the change of

capacitance before and after the target molecules bind to the receptors

immobilized on the electrode surface.

2

The difference in capacitance obtained before and after the binding of the

target molecule at the interface can however be negligibly small, making the

analysis of data using the conventional method of circuit fitting impossible [7].

This problem can be resolved using multivariate data analysis (MVDA) where

the whole spectra is taken into account to distinguish small changes of

capacitance [16-18].

In stripping voltammetry, the procedure involves two steps: i. pre-

concentration or accumulation of analytes on the electrode and ii. stripping or

dissolution of deposited analytes from the electrode. The method can be

employed in different modes based on the analyte of interest [11]. Square wave

anodic stripping voltammetry (SWASV) is the most sensitive technique and

has been mainly used for the detection of heavy metal ions [19, 20].

Besides the choice of transduction technique, another crucial factor for

detection in electrochemical sensor and biosensor development is the strategy

to immobilize a receptor on the electrode surface for recognizing and binding

target species [4, 6, 15, 21-23]. The way that the receptor attaches to the

surface plays an important role in enhancing the sensitivity, selectivity, and

reproducibility of the final sensor [22]. Modification of electrode with self-

assembled monolayer (SAM), gold-alkanethiol system, is the most attractive

approach that has been serving as a platform for attaching recognition

elements, such as antibodies, on the electrode surface [6, 22]. The system is

popular because it is easy to perform. One just needs to immerse a clean Au

electrode into a solution of alkanethiol with various terminal groups whereby

a well-organized monolayer is obtained [6, 24-29]. In the past decades, an

alternative surface modification system using diazonium salt has been

receiving great attention over alkanethiol-SAM [6, 22, 30-34]. This is due to

the short time needed for grafting, strong covalent bond that forms between

the aryl radical and electrode surface, and the stability of the grafted layer. The

grafted layer provides a wider potential window for performing

electrochemical measurements [6, 22, 32, 33].

3

1.1. Aims of the Thesis

The primary aim of this research was to develop electrochemical biosensor

systems that can be used for the detection of therapeutic drugs, using

electrochemical immittance spectroscopy (EIS) along with multivariate data

analysis. Methotrexate (MTX) was used as a model drug. The surface

modification method developed in this study was then extended to the

development of an electrochemical sensor for detection of Cd2+ and Pb2+ using

square wave anodic stripping voltammetry (SWASV).

1.2. Thesis Outline

In a previous work, the electrochemistry group at the department reported a

flow cell for the detection of MTX using nonfaradaic EIS and multivariate data

analysis [17]. The flow system consisted of two Au plate electrodes, that were

modified with anti-MTX through a cysteamine-SAM, between which a flow

chamber was sandwiched. The method showed promising results for detection

of trace levels of MTX in phosphate buffer solutions. In the present work, the

method was further developed for analysis of MTX in human blood serum,

paper I [35]. The concept was then extended further by using a different

surface modification protocol where diazonium salt is used for construction of

the biosensor, paper II [36]. A gold disk electrode, modified using 4-

carboxybenzenediazonium tetrafluoroborate (4-CBD), was used for detection

of methotrexate in a batch system. A detection of MTX was performed by EIS

measurements with and without application of redox probe, Fe(CN)63-/4-. The

Rct obtained from circuit fitting of impedance data of Fe(CN)63-/4- cannot be

used for the detection of MTX. However, the MTX can be determined when

EIS measurement, without redox probe, was coupled with the singular value

decomposition (SVD). The method provided a low limit of detection (7 × 10-12

M) which is as good as in paper I (5 × 10-12 M), and significant lower than other

electrochemical sensors. Nevertheless, the reproducibility and accuracy of the

system still were not satisfactory. This could be probably resulting from the

usage of high concentration of 4-CBD (2.5 mM) in electrode modification. At

such high concentrations a multilayer is possibly formed [37, 38] and this

4

could lead to a detrimental impact on sensor performance by causing

electrode fouling [39]. Thus, electrode modification using lower

concentrations of 4-CBD was examined on glassy carbon electrode (GCE)

using CV. The cyclic voltammogram apparently altered from what seen in high

concentration of diazonium. Thus the new insight into electrochemical

behaviour of 4-CBD on glassy carbon electrode (GCE) using cyclic

voltammetry is demonstrated in paper III (submitted manuscript). Finally,

a GCE, modified using 4-CBD combined with an in situ Bi3+ plating, was used

for detection of Pb2+ and Cd2+ using SWASV, paper IV (manuscript).

Although a number of studies have employed 4-CBD-based electrode

modification for analysis of heavy metals, deposition of an in situ Bi3+ on

modified electrodes has not been investigated earlier.

5

2. Background

2.1. Sensors and Biosensors

A sensor or chemical sensor is an analytical device that converts chemical

issue i.e concentration of sample composition into measurable analytical

signal [40, 41]. There are two main combination parts to form a sensor, a

recognition element (receptor) and a transducer. The receptor is an element

that interacts with a certain analyte to be detected in a high degree of

selectivity in which the interaction makes a chemical change. The transducer

is a physicochemical detector segment that transforms chemical response into

a quantifiable physical signal [40, 42].

Sensors can be classified according to their recognition elements and their

transduction methods. One can call a chemical sensor a biosensor when its

receptor is a biological sensing element (bioreceptor) such as cells, enzymes,

antibodies, nucleic acid, microorganism, tissues, etc [40]. These bioreceptors

contribute to numerous types of biosensors: whole-cell biosensor, enzymatic

biosensor, immunosensor (receptor is antibody), genosensor (receptor is

nucleic acid), DNA biosensor [43]. The transduction can be conducted in

various modes which have brought sensors and biosensors into three main

groups: electrochemical, optical, mass change based sensors and biosensors

[44-46].

In the development of a sensor, the first thing to be selected is a suitable

receptor molecule, suitable immobilization method, and transducer.

Electrochemical sensors and biosensors have attracted the attention of many

researchers because of their experimental simplicity, low cost, and remarkable

detection limits. They have been used in wide range of important applications

in the fields of clinical, industrial, environmental and agricultural analysis

[47-52]. The analytes to be detected can be organic, inorganic and biological

compounds [53].

6

2.1.1. Electrochemical Sensors and Biosensors

In electrochemical sensors, the electrode is utilized as a transduction element

[53, 54]. The sensor relies on the measurement of currents and/or voltages

which are given by a recognition event between the immobilized or chemically

modified receptor on the electrode and the target analytes [40, 55, 56].

Electrochemical transducers are mainly operated via cyclic voltammetry (CV),

amperometry, linear sweep voltammetry (LSV), differential pulse

voltammetry (DPV), square wave voltammetry (SWV), electrochemical

immittance/impedance spectroscopy (EIS), etc [44, 57]. Fig. 2.1 demonstrates

the electrochemical sensors and biosensors along with their measurable

signals obtained from different transduction techniques. The transduction

methods applied in this research are described in the forthcoming sections.

Fig. 2.1. Electrochemical sensor and its measurable signal.

2.1.1.1. Electrochemical Immittance Spectroscopy

Immittance spectroscopy is a concept that integrates the impedance (Z) and

admittance (Y), which is the reciprocal of impedance. The common basic

principle is focused on impedance spectroscopy.

7

Impedance is an opposition force to electrical current in a circuit that is

analogous to Ohm’s law of resistance. However, resistance is obtained in DC

circuit whereas impedance is observed in the AC circuit. For AC, a phase shift

may take place in time between the potential (E) and the current (I), Fig. 2.2,

which is different from DC, where there is no phase shift [58].

Fig. 2.2. The phase shift of current responds to the potential in AC circuit.

The impedance can be expressed according to Ohm’s law as:

𝑍 = 𝐸𝑡

𝐼𝑡 =

𝐸0 sin (𝜔𝑡)

𝐼0 sin (𝜔𝑡+ 𝛷)= 𝑍0

sin (𝜔𝑡)

sin (𝜔𝑡+ 𝛷) (2.1)

which can be represented as a complex number:

𝑍 = 𝐸𝑡

𝐼𝑡 = 𝑍0 exp(𝑗𝛷) = 𝑍0 (𝑐𝑜𝑠𝛷 + 𝑗𝑠𝑖𝑛𝛷) (2.2)

where Z is impedance (Ω), Z0 is a magnitude, Et is the potential at time t (V),

It is the current at time t (A), ф is a phase shift, ω is a radial frequency

(radians/s). ω = 2πf, where f is frequency (Hz).

In practice, to perform the electrochemical impedance spectroscopy (EIS)

experiment, a fixed sinusoidal voltage is applied (AC voltage) in a range of

t

E

t

I

Phase shift

8

frequency using a potentiostat that employs a three or two-electrode cell. The

cell consists of an electrolyte solution containing the compound under

investigation. Normally, the AC voltage applied to the cell is often small, 1 to

10 mV, in order to maintain the linearity between the current and voltage [59].

When the process is running, the current will flow through the electrochemical

cell. The current at each frequency is recorded by the potentiostat and

transformed by a software using equation 2.1 to calculate the real and

imaginary parts of the total impedance as presented in equation 2.2. The total

impedance in equation 2.2 can be rewritten as:

Z = Z′ + jZ″ (2.3)

where Z′ = Z0 cosф = |Z| cosф, Z″ = Z0 sinф = |Z| sinф, and j2 = -1

The EIS data can be presented by the Nyquist plot, which is the most common

way, or the Bode plot. The Nyquist plot (Fig. 2.3) is a plot of imaginary

impedance which is always negative (y-axis) against the real impedance (x-

axis). In the Bode plot (Fig. 2.4), the phase angle or the magnitude of

impedance (|Z|) is plotted against the frequency or logarithm of frequency. In

general, EIS data is analyzed by fitting an equivalent circuit, using a software,

to understand what has happened in the investigated system [59-61]. A variety

of electrical elements: resistor, capacitor, inductor, Warburg, and constant

phase element (CPE), can be assembled in series or/and parallel to provide a

complex equivalent circuit model [61, 62]. Table 2.1 shows the circuit elements

with their corresponding impedance [60-62].

In the electrochemical experiment, the impedance may occur from solution

resistance (Rs), double layer capacitance (Cdl), charge transfer resistance of

electrode (Rct) and Warburg (W) [59]. The Rs can be affected by concentration

of dissolved ions and conductance of the solution, the Rct reflects the electron

transfer ability of working electrode when the electrode interacts with

electrolyte, and the W represents the mass transfer toward the electrode

surface. The Cdl can be affected by the electrical charge of the layer on the

electrode. In some cases, when the layer is inhomogeneous, CPE is used

9

instead of Cdl [58]. Generally, CPE can represent various elements according

to its n value that ranges from 0 to 1 [61, 62]. When: n = 1; the CPE behaves as

pure capacitor, n = 0.5; the CPE is like a Warburg, and n = 0; the CPE

represents a pure resistor [60-63]. The combination of these elements in a

circuit model is known as Randle circuit [58, 59, 64], Fig.2.5.

EIS can be applied for the investigation of layers of chemicals or polymers,

coated on electrodes. The effects of different layers and coatings can be

recognized by charge transfer resistance (redox probe is used) or double layer

capacitance (no redox probe is used) in EIS measurement.

Fig. 2.3. Nyquist plot of impedance spectra, modified from [59].

10

-40

-35

-30

-25

-20

-15

-10

-5

0

10-1

100

101

102

103

104

105

106

0

250

500

750

1000

1250

1500

1750

2000

|Z|

(oh

m)

ph

as

e (

de

gr

ee

)

log f (Hz)

|Z|

Phase

Fig. 2.4. Bode plot of the impedance spectra.

Fig. 2.5. Randle circuit (circuit model) of the EIS spectrum in Fig. 2.3.

Table 2.1 Circuit elements with their impedance

Circuit element Symbol Impedance

Resistor R ZR = R

Inductor L ZL = jωL

Capacitor C ZC = 1/jωC

Warburg W Zw = 1/[Y0(jω)1/2]

CPE Q ZQ = 1/[Y0(jω)n]

Note: Y0 is the magnitude of Q for CPE and magnitude of W for Warburg.

11

2.1.1.2. Stripping Voltammetry

Stripping voltammetry is a method that requires a three-electrode system and

it comprises two steps:

Pre-concentration step: deposition of target compounds onto a working

electrode surface as amalgam or thin film. The deposition can be done

either by electrolytic deposition (anodic or cathodic process) or

adsorption (no electrolysis) to form a complex [11, 65]. This step

increases the sensitivity of the method.

Stripping step: is a dissolution of the deposited compounds from the

electrode surface into the solution by a potential scan. The resulting

current which is proportional to the concentration of the analyte is

recorded in this step. The potential scan can be operated using different

forms such as square wave (SW), differential pulse (DP), and linear

sweep (LS). A square wave is the most commonly used form due to its

high speed and its resulting current which is higher compared to DP and

LS [53].

There are different stripping techniques including anodic stripping

voltammetry (ASV), cathodic stripping voltammetry (CSV), and adsorptive

stripping voltammetry (AdSV) [53, 65]. Fig. 2.6 represents the steps involved

in ASV and CSV. The AdSV is similar to ASV and CSV, however, the pre-

concentration step is based on adsorption [53]. The stripping step can be

either oxidation or reduction depending on the nature or composition of the

electroactive complex (metal redox or ligand redox) [66].

Fig. 2.6. The process of anodic stripping voltammetry (ASV) and cathodic

stripping voltammetry (CSV).

12

The ASV in which the stripping step is executed by a square wave potential

scan is recognized as square wave anodic stripping voltammetry (SWASV) and

it is the most widely used technique for analysis of trace metal ions as well as

organic compounds. Practically, the method can be handled as demonstrated

in Fig. 2.7A, [11]. During the deposition (reduction step), the analytes are

accumulated onto the working electrode, which generally is done under

stirring of the solution at a constant potential for a certain time. During the

stripping step, the stirring is stopped and the potential is scanned to a positive

direction which causes the oxidation of the deposited species (for example,

metals) to the products (metal ions) that dissolve back into the solution. The

resulting voltammogram recorded during the stripping step is shown in Fig.

2.7B. The obtained peak current (ip) is proportional to the concentration of

analytes. The square wave form of potential scanned during the stripping step

is illustrated in Fig. 2.7C. The symmetrical square wave consists of the forward

pulse (the same direction as scan direction) and the reverse pulse which are

superimposed on a staircase potential [53, 67].

In the square wave scanning: τ (s) is the time for a single cycle of square wave

or one step of staircase, the pulse frequency (Hz) is 1/τ, Esw (V) is the pulse

amplitude, and Estep (V) is the step potential of staircase. The scan rate of the

square wave can be expressed as:

Scan rate (mV s-1) = Estep (mV)/ τ (s) (2.4)

The current is collected twice during each cycle, at the end of the forward (if)

and the reverse pulses (ir), then the difference between them is recorded

against the scanning potential. Since the two currents have opposite signs,

their difference is greater than either if or ir.

13

Fig. 2.7. (A) The SWASV process, (B) voltammogram obtained in the

stripping step of SWASV, and (C) square wave potential applied during the

stripping step of SWASV.

2.1.1.3. Cyclic Voltammetry

Cyclic voltammetry (CV) is an electrochemical technique that requires a three-

electrode setup: a working electrode, a reference electrode, and a counter

electrode. It has a broad application in the analysis of electroactive

compounds and is mostly used to study the reversibility and mechanisms of

redox reactions, electron transfer kinetics, and to investigate the follow-up

reaction of the product at the electrode [68-70].

In this technique, a start and an end potential are applied between the

reference electrode and the working electrode. The potential is scanned in

both forward and backward directions for certain number of cycles at a

constant rate. The obtained cyclic voltammogram provides important

parameters such as the anodic and the cathodic peak currents (Ipa and Ipc) as

well as their corresponding peak potentials (Epa and Epc). These parameters

14

can, for example, be used to characterize the redox behaviour of compounds,

elucidate the kinetics of electrode reactions [53].

2.1.2. Electrode Surface Modification

Electrode surface modification is one of the most important parameters in the

establishment of sensors. It is related to the incorporation of a receptor with

a transducer surface (electrode). This can be done through chemical

modification of the electrode surface to attach a layer that allows a proper

attachment of the recognition element (receptor). This step can limit the

reproducibility, sensitivity, and selectivity of the final sensor [6]. In

impedimetric immunosensors, the modified layer must not only permit the

target analyte to bind with the receptor (antibody) but it must also inhibit

nonspecific binding to the electrode surface [6, 71].

Various chemical reactions can be used for the immobilization of a recognition

element onto solid surfaces. Self-assembled monolayers (SAMs) using

alkanethiol (R-SH) seem to be the first option when the transducer is Au

electrode [72]. When a clean Au surface is immersed in the alkanethiol

solution, thiol groups chemisorb onto the Au surface to produce thiolate bond

(R-S-Au), and a monolayer forms several hours, eq. 2.5 [6, 73].

200 21 H/AuAuSRAuHSR nn (2.5)

The thiolate on Au covalently binds the bioreceptor through a crosslinker such

as glutaraldehyde or NHS-EDC, which is selected based on the functional

group of alkanethiol used.

In the last few years, electrode modification using diazonium organic salt has

become an interesting technique. This approach provides some advantages

over the use of alkanethiols: provides a diazonium with different functional

groups, gives stable surface layers, and requires less time for grafting on the

electrode [30, 31, 74, 75]. Additionally, the grafted surface can be used in a

wide potential window [6, 32-34, 76].

15

Diazonium is generally synthesized from the diazotization reaction of

aromatic amines with NaNO2 in a strong acid, tetrafluoroborate acid (HBF4)

[22]. The HBF4 is recommended as a replacement for HCl in the synthesis to

obtain tetrafluoroborate diazonium salt (ArN2+BF4

-) which is stable and with

no risk of explosion [30]. The half-life of diazonium solution in an aprotic or

acidic solvent is around five days, however it is not stable in aqueous solutions

when pH > 2 or 3 [30].

The modification of electrode surface using diazonium salt can be done by

electrografting of diazonium solution, either with pre-synthesis of the

diazonium salt or in situ formation of diazonium salt [22]. Electrografting via

pre-synthesis of the diazonium salt is conducted by dissolving the synthesized

diazonium salt in acetonitrile, containing tetrabutylammonium

tetrafluoroborate (NBu4BF4) which is commonly used as supporting

electrolyte, or dissolving in acidic aqueous solution, for instance, 0.1 M H2SO4

[30]. There are two steps in the electrografting: 1) electrochemical reduction

of the diazonium ion into aryl radicals and 2) covalently binding of the aryl

radical with the surface atom of a substrate [31, 77]. A multilayer is mostly

formed during this step, see scheme 2.1. 4-carboxybenzenediazonium

tetrafluoroborate (4-CBD) was used in this research.

16

R = p-substitution on aryl: in this work is –COOH, so the grafted layer is

carboxyphenyl (CP)

Scheme 2.1. Electrografting of diazonium salt.

2.2. Methotrexate (MTX)

Methotrexate, MTX (Fig. 2.8), is an antineoplastic drug used in the treatment

of different types of cancer; osteosarcoma, lung cancer, breast cancer,

leukemia, lymphoma, and rheumatoid arthritis [78-80]. The drug is normally

administered orally or parenterally with different dosages according to the

patient’s pharmacokinetic profile [17, 81].

Application of high dose drug can lead to tremendous toxicities such as low

white blood cell counts, lung and liver disease [82], and in some cases, life-

threatening [83, 84]. The level of MTX in plasma is usually investigated after

24, 48 and 72 h from the start of infusion [81]. The risk of toxicity can be

incidentally developed when the concentration of plasma MTX is above: 5-10,

1.0, and 0.1 µmol L-1 at 24, 48, and 72 h, respectively [83, 85]. It is thus

important to develop a system for monitoring MTX levels in patients’ serum

to avoid the side effects and to find a good prevention approach.

17

Fig. 2.8. The structure of methotrexate (MTX).

2.3. Analytical Methods for MTX Detection

Various methods have been implemented for the detection of MTX in

pharmaceutical formulations or biological and clinical samples. The most

commonly used methods are based on high performance liquid

chromatography (HPLC) combined with different types of detectors: UV or

UV-Vis [86-95], fluorescence [84, 96-98], mass spectrometry (MS) or tandem

mass spectrometry (MS/MS) [99-107], electrospray ionization mass

spectrometry (EIS-MS) [108], electrospray ionization tandem mass

spectrometry (EIS-MS/MS) [109, 110], and electrochemical detector (ECD)

[111]. Other techniques such as fluorimetry [112-114], UV spectrometry [115,

116], capillary electrophoresis (CE) [117-121], ion chromatography with

electrochemical detection [122] and immunoassay [123-126] have also been

used for the determination of MTX. Although the demands of accuracy,

selectivity, and sensitivity can be fulfilled with some of these methods, they

require a high amount of organic solvent which is environmentally unfriendly,

complicated sample preparation and highly skilled operator. Such

requirements increase cost and time of analysis [82]. Furthermore, some of

the mentioned facilities are not always available as a standard instrument in

hospital laboratories [87]. Therefore, the need for a selective, sensitive, fast

and inexpensive method for determination of MTX is still of great interest. For

these reasons, electrochemical sensors and biosensors have become the best

alternatives [82].

A variety of electrochemical sensors and electrochemical biosensors have been

developed, generally using square wave voltammetry (SWV) and differential

18

pulse voltammetry (DPV), for quantification of MTX as listed in Table 2.2.

Most of the reported methods depend on the faradaic electrochemical

behaviour of MTX (oxidation) with an application of potential (0.2 to 1.2 V vs

Ag/AgCl) at low pH. Immunosensors based on capacitive electrochemical

immittance spectroscopy are an alternative type of sensors in which a

measurement is done based on the change of capacitance, when the analyte

interacts with the receptor, without the need of a redox probe. However, there

have not been any innovations of such sensors for the detection of MTX. We

have in the current research, for the first time, reported a capacitive

immunosensor using electrochemical immittance spectroscopy for the

determination of MTX in blood serum. The electrochemical immittance

spectroscopy data was interpreted using multivariate data analysis.

19

Table 2.2 Electrochemical sensors and biosensors for the detection of MTX

Electrode receptor Electrochemical method Linear range (mol L-1) LOD (mol L-1) Reference

Au nanoparticles self-assembled on L-

cystein modified glassy carbon electrode

(nano-Au/LC/GCE)

Square wave anodic stripping voltammetry

(SWASV)

4.0 × 10-8 – 2.0 × 10-6 1.0 × 10-8 [127]

DNA Langmuir-Blodgete modified GCE

(DNA-LB/GCE)

SWASV 2.0 × 10-8 – 4.0 × 10-6 5.0 × 10-9 [128]

poly (L-lysine) modified GCE (pLL/GCE) Square wave voltammetry (SWV) 5.0 × 10-9 – 2.0 × 10-7 1.7 × 10-9 [80]

Silver solid amalgam electrode (AgSAE) Differential pulse adsorptive stripping

voltammetry (DPAdSV)

1.0 × 10-9 – 3.0 × 10-6 1.5 × 10-10 [78]

Au nanoparticles covered on multiwall

carbon nanotubes-ZnO composite

modified screen printed carbon electrode

(nano-Au/MWCNTs-ZnO/SPCE)

SWV 2.0 × 10-8 – 1.0 × 10-6 1.0 × 10-8 [129]

Multiwall carbon nanotubes-dihexadecyl

hydrogenphosphate modified GCE

(MWCNTs-DHP/GCE)

DPAdSV 5.0 × 10-8 – 5.0 × 10-6 3.3 × 10-8 [50]

20

Table 2.2 (Continued)

Electrode receptor Electrochemical method Linear range (mol L-1) LOD (mol L-1) Reference

Cobalt ferrite nanoparticles-reduced

graphene oxide composite in ionic liquid

modified GCE (CoFe2O4-rGO-IL/GCE)

Differential pulse voltammetry (DPV) 5.0 × 10−8 – 7.5 × 10−6 1.0 × 10-8 [130]

DNA immobilized on GCE (DNA/GCE) Potentiometric stripping analysis (PSA) 2.0 × 10−6 – 3.6 × 10−6 5.6 × 10-9 [131]

Hanging mercury drop electrode (HMDE) DPAdSV 2.5 × 10−8 – 25 × 10−8 2.0 × 10-9 [132]

Multiwall carbon nanotubes modified

screen printed electrode (MWCNTs/SPE)

SWV 5.0 × 10−7 – 1.0 × 10−4 1.0 × 10−7 [133]

Mercury meniscus modified silver solid

amalgam electrode (m-AgSAE)

DPV 5.5 × 10−8 – 5.5 × 10−7 1.8 × 10−9 [134]

Cyclodextrin-graphene hybride

nanosheets modified GCE

(CD-GNs/GCE)

DPV 1.0 × 10−7 – 1.0 × 10−6 2.0 × 10−8 [135]

Nitrogen-doped graphene decorated with

a palladium silver bimetallic alloy

modified GCE (PdAg-NG/GCE)

DPV 2.0 × 10−8 – 2.0 × 10−4 1.3 × 10−9 [136]

3D graphene-carbon nanotube network

modified GCE (3DG-CNT/GCE)

DPV 7.0 × 10-7 – 1.0 × 10-4 7.0 × 10-8 [82]

21

2.4. Heavy Metal Ions (Pb2+ and Cd2+)

Lead and cadmium are among the heavy metals which show no biological

effect in all organisms and are highly toxic to the environment and human

health even at trace levels [20, 137]. Lead originates from the burning of

gasoline containing lead, mining waste, paints, incinerator ash, and water

draining from lead pipes [65]. Naturally, lead level is very low in surface water

and groundwater, < 5 ppb. However in water and soil of some areas the level

of lead can be higher than 1000 ppb. Lead pollution is generated extensively

by human activities, which lead to a worldwide problem, and exposure to a

low level of lead can cause neurological, reproductive, cardiovascular, and

developmental disorders [138, 139]. Main sources of cadmium are

electroplating industries, mining industries, nickel-cadmium batteries,

production of polyvinyl chloride plastics, and paint productions [65, 140]. The

presence of cadmium in the environment is coupled to human activities i.e.

combustion of metal ore, waste burning, and usage of fossil fuels. Cadmium

can contaminate drinking water via the impurities in the zinc of coated pipes

and solders [141]. Exposure to cadmium through water, air, and food for a

long period can result in cancer and other harmful effects on humans,

including skeletal, renal, reproductive, cardiovascular, central and peripheral

nervous and respiratory systems [140].

World health organization (WHO) has listed these metals as the urgent

substances to be controlled and has limited the concentration of lead and

cadmium in drinking water to 10 µg L-1 and 3 µg L-1, respectively [141].

Therefore, there is a requirement for highly sensitive, rapid, accurate and

simple methods for the detection of these metals in environmental samples.

2.5. Methods of Detection for Pb2+ and Cd2+

Electrochemical methods based on stripping voltammetry (Table 2.3) has

been used for detection of heavy metals as an alternative to traditional

spectroscopic methods such as atomic absorption spectroscopy (AAS),

inductively coupled plasma optical emission spectrometry (ICP-OES) or

22

inductively coupled plasma mass spectrometry (ICP-MS) and flame atomic

absorption spectroscopy (FAAS) [20]. The stripping voltammetric methods

have attracted great interest of researchers because they are simple,

inexpensive, user and environmentally friendly, fast, and suitable for in-field

applications [20]. Furthermore they are superior in terms of sensitivity,

reproducibility, and accuracy for trace analysis of these metals [12].

In the past, stripping voltammetry was performed on mercury electrodes,

however, due to the serious toxicities of mercury, an alternative mercury-free

electrode which is more environmentally friendly has been highly suggested

[142, 143]. Lately, a number of publications, listed in Table 2.3, have shown

that bismuth film electrodes, carbon electrodes, and chemically modified

electrodes are alternative solutions [142]. Bismuth film electrodes have been

selected because of their low toxicity, ease of preparation, their ability to form

alloys with different metals which can offer the synergetic effect of enriching

target metals leading to better sensitivities [142, 143]. Researchers have put a

lot of effort to generate chemically modified electrodes for the detection of

heavy metals. Among those, 4-CBD is one of the interesting compounds which

has been used to modify the electrode [142, 144, 145].

In this study, bismuth is in situ deposited onto the 4-CBD-based modified

glassy carbon electrode for the detection of Pb2+ and Cd2+ using SWASV. This

is the first time the plating of bismuth coupled with grafted electrodes is

reported for the detection of heavy metals.

23

Table 2.3 Stripping voltammetry for the detection of heavy metals

Electrode receptor Electrochemical method Analytes Linear range

(µg L-1)

LOD of Cd2+

(µg L-1)

LOD of Pb2+

(µg L-1)

Reference

4-carboxyphenyl diazonium-

based modified glassy carbon

electrode (CP/GCE)

SWASV Cd2+, Pb2+ 0.5 – 50 0.13 0.20 [144]

4-carboxyphenly diazonium-

based modified screen printed

carbon electrode (CP/SPCE)

SWASV Pb2+ 1.6 – 15.5 no 0.25 [142]

4-carboxyphenyl diazonium-

based modified boron doped

diamond electrode (CP/BDDE)

SWASV Cd2+ 2.0 – 50 0.20 no [145]

in situ bismuth plated onto poly

(p-aminobenzene sulfonic acid)

modified GCE (Bi/p-

ABSA/GCE)

DPASV Cd2+, Zn2+

Pb2+

1.0 – 110

1.0 – 130

0.63 0.80 [143]

24

Table 2.3 (continued 1)

Electrode receptor Electrochemical method Analytes Linear range

(µg L-1)

LOD of Cd2+

(µg L-1)

LOD of Pb2+

(µg L-1)

Reference

in situ bismuth plated on

Fe2O3-graphene modified

glassy carbon electrode

(Bi/ Fe2O3-graphene/ GCE)

DPASV Cd2+, Pb2+,

Zn2+

1.0 – 100 0.08 0.07 [146]

in situ bismuth plated on

multiwalled carbon nanotube–

Nafion composite modified

glassy carbon electrode

(Bi/MWCNT–Nafion/GCE)

SWASV Cd2+, Pb2+ 1.0 – 45 0.02 0.03 [147]

- co-deposited in situ (bismuth

and antimony) on carbon paste

electrode (Bi-Sb/CPE)

SWASV Cd2+, Pb2+ 10 – 70

(in situ)

0.8 (in situ)

0.9 (in situ)

[148]

in situ bismuth modified boron-

doped diamond electrode

(Bi/BDDE)

DPASV

Cd2+, Pb2+,

Zn2+

not mention 1.34 7.04 [149]

25

Table 2.3 (continued 2)

Electrode receptor Electrochemical method Analytes Linear range

(µg L-1)

LOD of Cd2+

(µg L-1)

LOD of Pb2+

(µg L-1)

Reference

in situ Bi plated onto Co3O4

nanosheets modified indium tin

oxide (Bi/Co3O4/ITO)

DPASV Pb2+ 1 – 100 no 0.52 [150]

in situ bismuth deposited on

hybrid binder carbon paste

electrode (hybrid binder carbon

paste electrode= ionic liquid 1-

butyl-3-methylimidazolium

hexafluorophosphate,

[BMIM]PF6] + graphite +

paraffin oil) (Bi/ HCPE)

SWASV Cd2+, Pb2+ 1 – 90 0.12 0.25 [151]

in situ bismuth plated on

mesoporous carbon nitride/self-

doped polyaniline nanofiber

modified glassy carbon electrode

( Bi/MCN/SPAN/GCE)

SWASV Cd2+, Pb2+ 5 – 80 0.7 0.2 [152]

26

Table 2.3 (continued 3)

Electrode receptor Electrochemical method Analytes Linear range

(µg L-1)

LOD of Cd2+

(µg L-1)

LOD of Pb2+

(µg L-1)

Reference

ex situ indium and bismuth

deposited GCE (In-Bi/GCE)

SWASV Cd2+, Pb2+,

Zn2+

0 – 120 0.15 0.67 [153]

pencil drawn electrodes (PDEs) Linear sweep voltammetry, LSV

(poor signal)

DPASV

Pb2+ Two ranges:

83 – 330

330 – 915

no 9.5 [154]

magnetic chitosan nanoparticle

modified glassy carbon electrode

(Fe3O4-chitosan NPs/GCE)

SWASV Cd2+, Pb2+,

Cu2+, Hg2+

135 – 191 Cd2+

21 – 269 Pb2+

4.50 8.70 [23]

nanocarbon film formed by

unbalanced magnetron

sputtering on boron doped

silicon substrate (UBM

nanocarbon)

SWASV Cd2+, Pb2+ 0.5 – 500 Cd2+

5 – 500 Pb2+

0.5 5 [19]

27

3. Methods

3.1. Electrochemical Setup

Electrochemical measurements were performed with Modulab

electrochemical system, ECS (Solartron Analytical, UK). Paper I, a two-

electrode setup, consists of two Au plates with a flow chamber sandwiched

between them, was used for the EIS measurement, Fig. 3.1. A three-electrode

setup, comprising Au disk electrode or GCE working electrode, Ag/AgCl

reference electrode, and Pt counter electrode, was used to conduct CV, EIS

and SWASV measurements in a batch system, paper II, III and IV.

Fig. 3.1. The electrochemical set up of a flow system.

28

3.2. Electrode Cleaning

The working electrode was cleaned prior to all measurements. The Au plate

electrode was cleaned with a boiling solution of H2O2 (30%):NH4OH

(25%):H2O, (1:1:5, v/v/v), for 20 s, then rigorously washed with Milli-Q water

followed by ethanol [35]. The Au disk electrode was polished with 1.0 µm

alumina slurry and thoroughly washed with Milli-Q water. Then it was

cleaned electrochemically in 0.5 M H2SO4 by CV for 25 cycles from -0.3 to +1.5

V at 100 mV s-1, followed by rinsing with Milli-Q water, sonication in ethanol

for 5 min and ACN for 5 min [36]. The GCE was polished with Al2O3

suspension of decreasing particle size (1.0, 0.3 and 0.05 µm). The electrode

was thoroughly washed with Milli-Q water after polishing with each particle

size.

3.3. 4-Carboxybenzenediazonium Synthesis

The 4-CBD was synthesized following the method reported elsewhere [36].

Briefly, 35 mol of 4-aminobenzoic acid was dissolved in 10 mL of HBF4 (48%,

W/W) and 10 mL of H2O. The solution mixture was cooled in an ice bath with

continuous stirring. Another 5.0 mL of H2O was added to dissolve the white

precipitate that occurred during stirring. After 20 min of cooling in an ice

bath, a solution of 37 mol NaNO2 dissolved in a minimum amount of water,

was added dropwise. Thereby, a white precipitate of final product was formed.

The precipitate was vacuum filtered, washed with diethyl ether and dried

under vacuum at room temperature. The compound was thereafter stored in

a freezer at -20 °C.

3.4. Electrochemical Biosensor for Detection of MTX (paper I and II)

3.4.1 Electrode Surface Modification

Cleaned Au electrodes were used for the immobilization of bioreceptor, anti-

MTX antibody (Ab), following the procedure described below. The modified

electrodes were employed instantly for MTX detection.

29

Paper I [35]: the Ab immobilization was done through the attachment of

alkanethiol (cysteamine). The whole process was conducted by injection of the

desired solutions into a flow system, where the two cleaned Au electrodes were

mounted on a holder, with a flow cell sandwiched in between, Fig. 3.2. A

schematic process of immobilization is illustrated in Fig. 3.3. The electrode

obtained after each step is labeled as 1) Cys/Au, 2) Glu/Cys/Au, 3)

Ab/Glu/Cys/Au, and 4) Bp/Ab/Glu/Cys/Au. Where, Au = gold, Cys =

cysteamine, Glu = glutaraldehyde, Ab = antibody, and Bp = blocking polymer

(pTHMMAA).

Paper II [36]: the attachment of the Ab on Au disk electrode was done via

4-CBD and the process is summarized in Fig. 3.4. The electrode modification

after each step is marked as 1) CP/Au, 2) + 3) NHS-EDC/CP/Au, 4) Ab/NHS-

EDC/CP/Au, and 5) Bp/Ab/NHS-EDC/CP/Au. Where, CP = carboxyphenyl,

NHS = N-hydroxysuccinimide, EDC = N-(3-dimetylaminopropyl)-N’-

ethylcarbodimide.

Fig. 3.2. A flow system used in the immobilization process (paper I).

30

Fig. 3.3. The immobilization process of anti-MTX on Au electrodes via

cysteamine (reprinted from scheme 1, paper I [35]).

Fig. 3.4. The immobilization process of anti-MTX on Au electrodes via 4-

CBD (reprinted from scheme 2, paper II [36]).

31

3.4.2. Characterization

Paper I: the EIS and X-ray photoelectron spectroscopy (XPS) were

performed to control the electrode after each step of the modification process.

Paper II: the CV and infrared spectroscopy (IR) were run to check the

production of 4-CBD from the synthesis, whereas the modification steps of

the electrode were monitored by CV and EIS.

3.4.2.1. Electrochemical Immittance Spectroscopy

The EIS was carried out on Au electrode, before and after each electrode

modification step.

Paper I: the measurement was made in PhB (pH = 7) in a frequency range of

0.5 to 100000 Hz, AC amplitude of 10 mV, DC level of 0 V vs reference and 10

data points/decade. The solution was pumped into the flow cell at a rate of

185 µL min-1 around 5 min to avoid the bubble then stopped for the

measurement.

Paper II: the EIS was conducted in 5.0 mM Fe(CN)63-/4- containing 100 mM

KCl in the frequency range: 0.5 to 65000 Hz, AC amplitude: 10 mV, DC level:

0.235 V vs reference (open circuit potential of Fe(CN)63-/4-) and 10 data

points/decade.

3.4.2.2. X-ray Photoelectron Spectroscopy (paper I)

The XPS was used to investigate the element compositions that are present on

the electrode surface before and after the modification process. The spectra

were recorded with a Kratos Axis Ultra DLD electron spectrometer using

monochromated Al Kα source operated at 150 W. The survey spectra were

collected from 1100 to 0 eV at pass energy of 160 eV. A high-resolution spectra

for O 1s, N 1s, C 1s, S 2p, and Au 4f were collected at pass energy of 20 eV. The

surface potential was stabilized by the spectrometer charge neutralization

system. The binding energy (BE) scale was referenced to the Au 4f7/2

photoelectron line, set at 84.0 eV. The spectra were achieved by the Kratos

software.

32

3.4.2.3. Cyclic Voltammetry (paper II)

The synthesized 4-CBD was characterized by running CV, +0.5 to -0.3 V at

100 mV s-1 , in 2.5 mM 4-CBD containing 50 mM of NBu4BF4 in ACN.

The modification step was tracked by scanning CV on bare Au electrode and

after each modification step in 5.0 mM of Fe(CN)63-/4- containing 100 mM KCl

from +0.6 to -0.5 at 100 mV s-1.

3.4.2.4. Infrared Spectroscopy (paper II)

The synthesized 4-CBD was characterized using IR spectrometer (AlPHA-T

BRUKER IR), in comparison with 4-aminobenzoic acid (parent compound),

in the range of 600-4000 cm-1.

3.4.3. Detection of MTX using EIS

Paper I: Analysis of MTX was performed in blood serum of a healthy person

which was provided by the University Hospital of Umea (NUS). The spiked

serum with different concentrations of MTX and the control (the serum

without MTX spike) were prepared as described in detail in paper I. The EIS

measurements were run in the order shown in Table 1, paper I and each

sample was pumped into the flow cell at 185 µL min-1 around 5 min then

stopped for measurement. The EIS parameters were the same as in the

characterization step: frequency range of 0.5 to 100000 Hz, AC amplitude of

10 mV, DC level of 0 V vs reference and 10 data points/decade.

Paper II: the MTX in PhB (pH = 7) was analyzed by EIS in two ways: (1)

direct measurement, and (2) measurement through Fe(CN)63-/4-. (1), the order

of measurement is presented in Table 1, paper II and the EIS parameters

were: frequency range of 0.5 to 65000 Hz, AC amplitude of 10 mV, DC level

of 0 V vs reference and 10 data points/decade. (2), the electrode was first

exposed for 30 min to a sample solution in Table 1, paper II, then washed

with water and transferred to a solution of 5.0 mM Fe(CN)63-/4- containing 100

mM KCl for EIS measurement. The EIS parameters were the same as in (1)

but the DC level was 0.235 V vs reference.

33

3.4.4. Multivariate Data Analysis

The immittance data are collected as:

Z = Z' + jZ'' (3.1)

where Z is the total impedance, Z' and Z'' are the real and imaginary and j2 =

-1. Z is a matrix of complex numbers of I measurements and K frequencies.

Paper I: frequency (100000-0.5 Hz), K = 55 and I = 39 (13 samples were

measured in 3 replicates each).

Paper II: frequency (65000-0.5 Hz), K = 53 and I = 30 (10 samples were

measured in 3 replicates each).

Z is usually transformed to complex capacitance C:

C = 1/[jωZ]= C'+jC'' (3.2)

where, ω = 2πf and f is the vector of the frequencies in Hz, C' is related to

dielectric constant and C'' is related to dielectric loss.

As explained earlier in [13, 16], singular value decomposition (SVD) of the

matrix C gives:

C = UASAVA* + E = TAVA* + E (3.3)

UA is a unitary (complex numbers) matrix for A component, VA* is a conjugate

transpose of a unitary matrix from A component. SA is a diagonal matrix of

real singular values. TA = UASA is sometimes easier to work with, paper II [36].

E is a residual matrix of irrelevant data or noise [13]. The total sum of squares

of C can be expressed as:

SStot = s12 + s2

2 +……..+ s2(E) (3.4)

Where s1, s2 are the singular value of first component and second component

and s2(E) is the residual variance (the dots stand for more components if

needed).

Multivariate calibration model was established to make a prediction for the

MTX concentration and is calculated using:

34

Paper I y = [1 U]b + f (3.5)

Paper II y = [1 T]b + f (3.6)

y is the vector for the known MTX concentration, U and T come from Eq. 3.3

with all values converted to real, 1 is the all-ones vector, b is the regression

coefficient and f is the vector of residuals.

In paper II, prediction tests were done using the capacitance spectra of

test/unknown samples, Cunk, which were not included in the calibration.

Tunk = CunkVA (3.7)

VA comes from Eq. 3.3 (for the calibration set)

ypred = [1 Tunk]b (3.8)

Tunk has all values converted to real just as T in Eq. 3.6

All calculations were performed in Matlab (Mathworks Inc, Natick MA)

3.5. Electrografting of 4-CBD on GCE (paper III)

3.5.1. Electrochemical Behaviour of 4-CBD on GCE

The electrochemical behaviour of 4-CBD on GCE was investigated using two

concentrations, 2.5 mM and 0.15 mM, by CV scanning from 0.7 to -0.7 V at

0.1 V s-1.

3.5.2. Electrochemical Oxidative Cleaning of Grafted Surface

The possibility of electrochemical oxidative cleaning of the grafted layer on

electrode surface was investigated. The 4-carboxyphenly (CP) was grafted on

bare GCE by CV scanning in 0.15 mM 4-CBD for 10 cycles between 0.7 and -

0.7 V at 0.1 V s-1. The grafted GCE (CP/GCE) was then rigorously washed with

Milli-Q water and sonicated with ACN for 5 min to eliminate adsorbed residue

on the electrode surface. Thereafter, the CP/GCE was used to run CV in a

freshly prepared solution of 0.15 mM 4-CBD. At this time, the potential was

scanned starting from more positive potentials, i.e. 1.2, 1.3 and 1.4 V to the

35

same cathodic limit (-0.7 V) using the same scan rate, to see whether the

extended anodic potential could burn the grafted layer off.

Another test was done with a thinner film of CP/GCE, in which the CP layer

was grafted using a single cycle. The grafting was conducted with 0.15 mM 4-

CBD by scanning 3 cycles between +1.4 to -0.7 V at scan rate of 0.1 V s-1. For

the single scan, the grafting is expected to occur in the forward scan (+1.4 to -

0.7 V). The reverse scan is supposed to remove the layer, grafted during the

forward scan, through oxidative cleaning. Two more cycles (2nd and 3rd) were

performed to check whether the layer grafted during the 1st cycle is still there

or removed. This can be checked by comparing the reduction peak potential

obtained in the 1st cycle with that obtained in the 2nd and 3rd cycles.

3.5.3. Effect of 4-CBD Concentration and Scan Rate on Grafting

The 4-CBD concentration and the scan rate play important roles on the

electrode grafting, so their impact was studied. Various concentrations of 4-

CBD, 0.05 to 0.30 mM, were used to graft on GCE by CV, scanned from +0.7

to -0.7 V at 0.1 V s-1. The scan rate was varied in the range of 0.010 to 2.5 V s-

1 during the CV measurement of 0.15 mM 4-CBD on GCE, ran from +1.4 to -

0.7 V.

3.6. Electrochemical Sensor for Detection of Metal Ions (paper IV)

A GCE was modified using 4-CBD to generate carboxyphenyl-GCE (CP/GCE)

and coupled with an in situ electrodeposition of Bi for the detection of Pb2+

and Cd2+.

3.6.1. Preparation of Carboxyphenly-GCE (CP/GCE)

A cleaned GCE was electrografted by running CV in 2.5 mM 4-CBD solution

for 5 cycles from +0.7 to -0.7 V at 100 mV s-1 to deposit the carboxyphenyl

(CP) layer on the electrode. The modified GCE (CP/GCE) was thereafter

sonicated in ACN for 5.0 min and used for further measurements.

36

3.6.2. Detection of Pb2+ and Cd2+

Square wave anodic stripping voltammetry, SWASV, was used for the

detection of Cd2+ and Pb2+. The measurement was done on the CP/GCE in

standard solutions of Cd2+ and Pb2+ containing 1.0 mg L-1 Bi3+ in 0.1 M Ac

buffer (pH = 4.5). The CP/GCE was immersed in the solution for 15 s, to allow

the adsorption of Pb2+, Cd2+, and Bi3+ to the electrode, before measurement

with square wave voltammetry (SWV). The SWV measurement was made

from -1.0 V to +0.0 V under the optimized conditions: pulse amplitude (25

mV), step potential (1 mV), pulse frequency (25 Hz) and integration period

(20 % of pulse width). No pre-electroreduction step (preconcentration) of the

adsorbed metal ions was used. However since the scan starts at -1.0V the

metal ions will be instantaneously reduced at this potential and stripped back

to the solution when the scan reaches the oxidation potential of Cd ( ̴ -0.8 V),

Pb ( ̴ -0.6 V), and Bi ( ̴ -0.2 V). The Bi3+ was added to the analyte solution for

in situ electrochemical deposition (electroreduction) to get the Bi on the

CP/GCE which is referred as Bi/CP/GCE.

3.6.3. Detection of Pb2+ and Cd2+ in Tap Water

Tap water, taken from one of the water taps at the laboratory, was used as a

sample. Prior to analysis, 0.1 M HAc and 0.1 M NaAc were added to the sample

and adjusted to pH 4.5. An interim solution containing 100 mg L-1 of both Cd2+

and Pb2+ was used for the standard addition. Appropriate volumes of the

interim solution were used to get the added concentrations in the sample: 70,

80, 90, 100, 110 and 120 µg L-1 with respect to both Cd2+ and Pb2. The

concentration of Bi3+ was 1.0 mg L-1 in all solutions. A calibration curve was

prepared using the 70, 90, and 120 µg L-1 solutions. The remaining

concentrations, 80, 100, and 110 µg L-1 were used for recovery test.

37

4. Results and Discussion

4.1. Paper I

4.1.1. Characterization of Electrode Modification steps

The stepwise modification of the Au electrode was characterized using the

double layer capacitance (Cdl), obtained from EIS spectra (Fig. 2A, paper I),

which are evaluated graphically by C"max × 2 (Table 2, paper I).

The Cdl of bare Au electrode, Cys/Au and Glu/Cys/Au were obviously

different. The Cdl of bare Au was increased when modified with cysteamine,

Cys/Au, which is due to the increase of charge storage capacity of the Au-

bound cysteamine by the protonation of amino group (NH3+-(CH2)2-S-). The

Cdl on Glu/Cys/Au was lower because of the growth in thickness when Glu is

attached to cysteamine. This indicated that the reaction took place after each

modification step yielding the new surface as expected. However, the Cdl of

Ab/Glu/Cys/Au and Bp/Ab/Glu/Cys/Au did not show any differences

comparing to Glu/Cys/Au. To confirm the attachment of Ab with Glu on the

electrode surface, XPS was used. The peaks obtained at 287.4 eV (C 1s), 399.2

eV (N 1s) and 530.8 eV (O 1s) correspond to the amide function (-N-C=O) and

are distinctive for the presence of protein. Another peak of N 1s observed at

397.7 eV corresponds to N=C function. The occurrence of these functional

groups is a firm expression of Ab bound on the surface since the protein is the

main component of the antibody and N=C is the bond generated in the

reaction between Glu and Ab.

4.1.2. Detection of MTX using EIS

The measurement of MTX using EIS gave the data matrix C (39 × 55) and the

data were analyzed without pretreatment using SVD. The score plot of the 1st

component of the real and imaginary scores, u (1), demonstrates a cluster of

the MTX scores which is separated from the blank and the control (unspiked

serum) scores, Fig.7A, paper I. This indicated that there is specific

recognition of the methotrexate by the antibody. The score plot showed a time

drift in the response that was orthogonal to the concentration gradient and

38

this allows a calibration of the MTX to be constructed. To figure out the

frequency range that is responsible for the separation, the EIS data was

analyzed by dividing the whole frequency range into two regimes: 100 to

100000 Hz (Fig. 7B, paper I) and 0.5 to 100Hz (Fig. 7C, paper I). A clear

separation of MTX concentrations was observed in the range of 0.5 to 100 Hz.

This leads to the conclusion that the binding of MTX to the antibody occurs at

low frequencies, leading to the relaxations observed at these frequencies in

the immittance spectra (Fig. 6B, paper I). The calibration curve of MTX,

obtained in this frequency range, showed linear relationship between the

predicted and measured concentration of MTX (log scale) in the range of 2.73

× 10-12 to 2.73 × 10-4 mol L-1 (R2 = 0.9845) with limit of detection (LOD) of 5

× 10-12 mol L-1.

4.2. Paper II

4.2.1. Characterization of Electrode Modification Steps

The peak currents in the CVs of Fe(CN)63-/4- run on bare Au and modified Au

electrodes for the different steps (Fig. 2A, paper II), were in agreement with

the Rct calculated from EIS measurements (Fig. 2B and Table 2, paperII).

The results obtained from CV and EIS are an indication that the antibody was

immobilized on Au electrode.

4.2.2. Detection of MTX using EIS of Redox Probe, Fe(CN)63-/4-

The results were based on Rct obtained from circuit fitting of the immittance

spectra of Fe(CN)63-/4-, corresponding to MTX and blank (PhB) solutions. The

Rct continuously increased irrespective of exposure to PhB or MTX solution,

Fig 4.1. Therefore, the increase in Rct could not be used for explaining the

interaction of MTX with antibody on the electrode surface. The increase in Rct

was just due to time drift, thus the obtained Rct could not be used for detection

of MTX.

39

80

100

120

140

160

180

200

220

0 5 10 15 20 25 30 35

Rct /

oh

m

Blank and MTX measurements

PhB 1

MTX 1

PhB 2

MTX 2

PhB 3

MTX 3

PhB 4

MTX 4

PhB 5

MTX 5

Fig. 4.1. The plot of Rct for all five blanks (PhB) and all MTX concentrations

(reprinted from Fig. S2-B, [36]).

4.2.3. Detection of MTX using EIS without Redox Probe

The EIS data obtained from the direct measurement of MTX and PhB without

a redox probe were analyzed using SVD. The data of capacitance was arranged

into a matrix C (30 × 53), mean-centered and uv scaled before the SVD

calculation. There was a clear separation between MTX scores and PhB scores

for both components (1st and 2nd), Fig. 4.2A and 4.2B. The separation between

the MTX and PhB is explained by the loading plot (more details in paper II),

briefly, it was due to the interaction of MTX with the Ab at around 100 Hz (the

relaxation frequency). Figure 4.2A shows that the concentration dependence

of the MTX might interfere with time dependance because the trend of the

scores is in the same direction. This was corrected for by subtracting the

capacitance data obtained for the blank from that obtained for the MTX,

thereafter, used for making a regression.

The regression model was linear in the range of 3 × 10-12 to 3 × 10-4 mol L-1

with LOD of 7 × 10-12 mol L-1. The regression model with the data matrix of

MTX, (15 × 53), was validated by dividing it into a calibration set (10 × 53)

40

and a test set (5 × 53). For the calibration, a root mean square error of

calibration, RMSEC= 0.37 and R2 = 0.9775 were obtained. The small value of

RMSEC and high R2 expresses that the model was good. This calibration was

then used to predict the MTX concentration in a test set and provided a root

mean square error of prediction, RMSEP = 1.18 and R2 = 0.8692. The

predicted log[MTX] provided 2.5 to 23 % of relative error compared to the

true log[MTX]. The relative error is quite high, however, the model is still

reliable for the prediction, because all 15 measurements of blank (Table 1,

paper II) were found to be outside the MTX concentration range.

-0.10

-0.05

0.00

0.05

0.10

0.15

-0.40 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 0.40

Im u

(1)

Re u (1)

(A)

Time

Conc.

123

456

789

1011

12

131415

1617

18

192021

222324

2526

27

PhB

2829

30

-0.10

-0.05

0.00

0.05

0.10

0.15

-0.40 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

Im u

(2)

Re u (2)

(B)

1 2 3

4 5 6

7-9

10

11

12

13-15

16-18

19-21

22-24

25-27

2829

30

Fig. 4.2. The 1st and 2nd components obtained from SVD explained 72.9% and

26.8% of the data variation, respectively. (A) score plot for the 1st component

of the real and imaginary, (B) score plot for the 2nd component of the real and

imaginary. PhB = black cross, MTX 1 (3 × 10-12 mol L-1) = squre filled green,

MTX 2 (3 × 10-10 mol L-1) = pink triangle, MTX 3 (3 × 10-8 mol L-1) = red circle,

MTX 4 (3 × 10-6 mol L-1) = violet strip square and MTX 5 (3 × 10-4 mol L-1) =

blue cross square. The number labeled in score plots identify each

measurement. (Reprinted from Fig. 4A and 4C, [36], with adjustment of PBS

to PhB).

41

4.3. Paper III

4.3.1. Electrochemical Behaviour of 4-CBD

The CV of 4-CBD at 2.5 mM and 0.15 mM on GCE provided different results,

Fig. 4.3 A and B. The CVs of 2.5 mM 4-CBD, Fig. 4.3 A, resulted in a very broad

reduction peak which consisted of two peaks at 0.20 V (peak 1) and -0.01 V

(peak 2), during the 1st cycle. The peak disappeared in the successive cycles,

2nd and 3rd. However, for the lower concentration of 4-CBD, 0.15 mM (Fig. 4.3

B), a single and sharp reduction peak at 0.3 V was observed in the 1st cycle. In

the subsequent cycles (2nd to 22nd) the peak current did not decrease

drastically as was the case for the higher concentration. Instead, the peak

potential shifted continuously and gradually to more negative values until the

peak was no longer clearly identifiable from 23rd to 26th cycle. The different

results obtained for both concentrations could be explained based on the

number of monolayers provided during grafting.

The number of monolayers calculated from CVs is based on the assumption

that all carboxyphenyl (CP) radicals generated from the reduction of 4-CBD

bind to the electrode surface (overestimation). At high concentration, a

multilayer (100 monolayers) is formed during the 1st CV scan, whereas at low

concentration, 22 cycles (1st to 22nd) are required to form a multilayer (35

monolayers). This indicates that at low concentration the number of

monolayers are gradually built to result in a multilayer. The number of

reduction peaks depend on the number of monolayers deposited on the

electrode. See more details in manuscript III.

42

Fig. 4.3. CV of blank, 50 mM NBu4BF4 + ACN (blue line) and 4-CBD: (A)

2.5 mM for 3 cycles (black line); (B) 0.15 mM for 26 cycles (black and colored

lines): 1st cycle (magenta), 10th cycle (cyan), 23rd cycle (red) and 26th cycle

(green). Scan rate of 0.1 V s-1; arrow shows the direction of scan.

4.3.2. Electrochemical Oxidative Cleaning of the Grafted Surface

The thicker film of CP grafted on GCE by scanning 0.15 mM 4-CBD for 10

cycles from +0.7 to -0.7 V was not possible to remove when the potential of

the anodic scan was extended to 1.2, 1.3 and 1.4 V, Fig. 4.4A. This was verified

by the change of reduction potential to more negative values when the scan

was made to 1.2, 1.3 and 1.4 V. The potentials shifted to more negative values

are due to high charge transfer resistance of the grafted layer. However, the

thin layers of CP grafted on GCE by scanning for 1 cycle from +1.4 to -0.7 V

could be cleaned in the reverse scan -0.7 to +1.4 V. This is indicated by the

appearance of a peak at around 1.0 V in the anodic scan which corresponds to

the oxidation of carbon surface, Fig. 4.4B. Additionally, the reduction peak

potential obtained in the 2nd and 3rd cycle did not shift compare to the

potential peak in the 1st cycle.

43

Fig. 4.4. CVs of 0.15 mM 4-CBD: (A) 1st (black line) and 10th (dot black line)

cycle, scan range 0.7 to -0.7 V; 1.2 to -0.7 V (cyan line); 1.3 to -0.7 V (green

line); 1.4 to -0.7 V (red line); (B) Grafting of GCE from 1.4 to -0.7 V: 1st (black),

2nd (red), and 3rd cycle (green). Scan rate 0.1 V s-1. Blue lines in (A) and (B)

represent blank and arrows show the direction of scan.

4.3.3. Effect of 4-CBD Concentration and Scan Rate on Grafting

The number of monolayers can be controlled by varying the concentration of

4-CBD and the scan rate of CV. The number of monolayers increased linearly

from 0.9 to 4.3 when the concentration was increased from 0.05 to 0.30 mM

at a scan rate of 0.1 V s-1, Fig. 4.5A. The CP radicals are formed in high amount

and bind on the surface when increasing the 4-CBD concentration. In

contradiction to 4-CBD concentration, the number of monolayers decreased

from 6.0 to 0.4 when the scan rate was raised from 0.01 to 2.5 V s-1 keeping

the concentration of 4-CBD at 0.15 mM, Fig. 4.5B. This is attributed to the

shorter time available for grafting (840 to 210 ms) at higher scan rates.

44

Fig. 4.5. (A) The number of monolayers vs 4-CBD concentration, and (B) the

number of monolayer vs scan rate.

4.4. Paper IV

4.4.1. Procedure for Measurement of Cd2+ and Pb2+ and the Performance of Different Electrodes

In a preliminary test, SWASV measurements were done with 500 µg L-1 of both

Cd2+ and Pb2+ in 0.1 M Ac buffer solution (pH = 4.5) using 2 procedures; 1: (i)

immersion of electrode in analyte solution for 15 s, (ii) pre-electroreduction of

the adsorbed metals on the electrode at -1.2 V for 120 s, and (iii) stripping by

SWV; and 2: (i) immersion of electrode in analyte solution for 15 s, and (ii)

stripping by SWV.

The reactions suggested to be involved in procedure 1 and procedure 2 are

presented in mechanism 1 and mechanism 2, respectively.

Mechanism 1:

(i) Adsorption: nAr-COO- + Mn+ (Ar-COO)nM

(ii) Pre-electroreduction: (Ar-COO)nM + ne- nAr-COO- + M0

(iii) SWV stripping: M0 Mn+ + ne- (re-oxidation)

Mechanism 2:

(i) Adsorption: nAr-COO- + Mn+ (Ar-COO)nM

(ii) SWV stripping:

45

(Ar-COO)nM + ne- nAr-COO- + M0 (at the starting potential of

the scan)

M0 Mn+ + ne- (re-oxidation or stripping)

The initial setup of SWV used in the measurement were: pulse frequency (25

Hz), potential step (4 mV), pulse amplitude (25 mV), and integration period

(10 % of pulse width). Four types of electrodes (GCE, Bi/GCE, CP/GCE and

Bi/CP/GCE) were tested for these measurements to check their performance

in the detection of Cd2+ and Pb2+. The CP was grafted on GCE by running CV

in 2.5 mM 4-CBD for 10 cycles, from +0.7 to -0.7 V at 100 mV s-1. The

electrodes modified with Bi (Bi/GCE or Bi/CP/GCE) were obtained during

SWASV measurement on GCE or CP/GCE by adding 1.0 mg L-1 Bi3+ to the

analyte solution for in-situ deposition of Bi.

Fig. 4.6A, and 4.6B represent the square wave voltammograms of Cd2+ and

Pb2+ with different electrodes measured following procedure 1 and procedure

2, respectively. With both procedures, Bi/CP/GCE has the highest

performance compared to the other electrodes. This effect is attributed to the

electrostatic interaction between negative charge of -COO- on CP with positive

charge of metal ions and this promotes the electroreduction of metal ions

[144]. Furthermore, the Bi deposited on the surface during electroreduction

facilitates the reduction of the Cd2+ and Pb2+ by forming a fused alloy which

enhances the accumulation of Pb and Cd [150]. However, the currents for Pb2+

and Cd2+ obtained from Bi/CP/GCE in procedure 2 are higher than procedure

1, Fig. 4.6C. Moreover, the variation (% RSD, n = 6) of the currents provided

by procedure 2, 3 % (Cd2+) and 2 % (Pb2+), is less compared to % RSD of the

currents given by procedure 1, 13 % (Cd2+) and 8 % (Pb2+). The higher

fluctuation of currents obtained by using procedure 1 could probably be due

to electrostatic repulsion between the free negative charge of -COO- on the

electrode, and the high negative potential applied during the pre-

electroreduction (mechanism 1). The repulsion is more pronounced when

higher negative potential is applied for a longer time (procedure 1,

mechanism 1) compared to lower negative potential with shorter time

46

(procedure 2, mechanism 2) [155]. Therefore, procedure 2 was selected for

further study.

Fig. 4.6. Square wave voltammograms of 500 µg L-1 of Pb2+ and Cd2+ in 0.1

M Ac buffer (pH = 4.5), on bare GCE, CP/GCE, Bi/GCE and Bi/CP/GCE,

measured by: (A) procedure 1, with pre-electroreduction step (-1.2 V for 120

s), and (B) procedure 2, without pre-electroreduction step. Concentration of

Bi3+: 1.0 mg L-1; grafting with 2.5 mM 4-CBD using CV: 10 cycles, from +0.7

to -0.7 V at 100 mV s-1; and SWV initial setup parameters: potential scanning

(-1.0 to 0.0 V), pulse frequency (25 Hz), potential step (4 mV), pulse

amplitude (25 mV), and integration period (10 % of pulse width). (C)

Responses for Cd2+ and Pb2+ on Bi/CP/GCE using procedure 1 and procedure

2. Error bars: standard deviation (n =6).

47

4.4.2. Optimization of Relevant Parameters

To find the optimal conditions for the detection of Cd2+ and Pb2+, the

parameters for the experimental measurement were varied. The target

solution used in the optimization process was a standard solution of 500 µg L-

1 Cd2+ and Pb2+ containing 1.0 mg mL-1 Bi3+ in a 0.1 M Ac buffer solution (pH

= 4.5). The optimized experimental conditions are summarized in Table 4.1.

Table 4.1. Optimized experimental conditions.

Optimization Parameter Initial

condition

Studied

range

Optimal

value

SWV set up

Integration

period (% of

pulse width)

10 10-20 20

Step potential

(mV)

4 1-10 1

Pulse amplitude

(mV)

25 10-35 25

Pulse frequency

(Hz)

25 25-65 25

CP/GCE

preparation

Number of

cycles

10 1-12 5

4-CBD

concentration

(mM)

2.5 1.0-4.0 2.5

Analytical

conditions

Bi3+

concentration

(mg L-1)

1.0 0.1-2.0 1.0

pH 4.5 3.5-5.5 4.5

Immersion time

(s)

15 15-600 15

4.4.3. Analytical Performance

The calibration curves obtained for both Pb2+ and Cd2+, using the optimal

conditions listed in Table 4.1, are displayed in Fig 5B, paper IV. The linear

48

ranges for Cd2+ and Pb2+ were 50-500 µg L-1 (R2 = 0.9945) and 25-500 µg L-1

(R2 = 0.9928), respectively. The method provided LOD (S/N = 3) of 25 µg L−1

for Cd2+ and 10 µg L−1 for Pb2+.

The repeatability of measurement was checked by making six successive

measurements using the same CP/GCE electrode in standard solutions

containing both Pb2+ and Cd2+ at three concentration levels: 50, 300, and 500

µg L−1 in 0.1 M Ac (pH =4.5) in the presence of 1.0 mg L-1 of Bi3+. For the

reproducibility study, the measurements of Pb2+ and Cd2+ in the same three

standard solutions were repeated six times using six different CP/GCE

preparations. The RSDs of current for Cd2+ and Pb2+ at each concentration on

the same electrode and on six electrodes, Table 2, paper IV, are less than

the acceptable limits of %RSD calculated according to Association of Official

Analytical Chemists (AOAC) guideline [156]. These results suggest that

repeatable and reproducible measurements can be made using the

Bi/CP/GCE electrode.

4.4.4. Determination of Cd2+ and Pb2+ in Tap Water

To evaluate the practical use of the proposed method, Cd2+ and Pb2+ were

determined in tap water using Bi/CP/GCE. Cadmium and lead ions could not

be detected in the tap water, probably due to their presence at very low

concentrations, lower than LOD of the method. This was the case even with

ICP-OES. The accuracies of both methods for the determination of Pb2+ and

Cd2+ were evaluated using recovery test. As reported in Table 3, paper IV,

the recoveries for both Cd2+ and Pb2+ varied from 96.2 ± 5.4 to 104 ± 2%

(proposed method) and 83.9 ± 7.0 to 102 ± 1% (ICP-OES). These recoveries

are acceptable according to AOAC guideline [156], which sets the acceptable

range to 80 % - 110 %, when the concentration is bigger or equal to 100 µg L-

1. This demonstrates that our proposed method is applicable to detect the

metal ions in real water samples with comparable results to ICP-OES.

49

5. Conclusions

The cutting-edge of electrochemical biosensors based on non-faradaic EIS

and SVD were presented for the detection of MTX, paper I and II. The

biosensor displayed orthogonal behaviour of the MTX concentration gradient

and time drift in the score plot for the whole frequency range, 0.5 to 105 Hz,

(paper I). A lower frequency range 0.5 - 100 Hz was sufficient for good

separation in the score plot and for making regression model. The regression

gave linear relationship in the range of 2.73 × 10-4 to 2.73 × 10-12 mol L-1 with

a LOD of 5 × 10-12 mol L-1, which was the lowest compared to other reported

methods. However, it was evident that the surface preparation of the electrode

needed to be improved to get more reproducible results. In paper II, where

the use of a redox probe and direct measurement for the detection of MTX

was compared, the large time drift that was observed during the

measurements made the detection of MTX impossible when the redox probe

is used. For the direct measurement, without using redox probe, the MTX

could be detected by evaluating the EIS data using SVD. The regression of the

corrected data of MTX was linear in the range of 3 × 10-12 to 3 × 10-4 mol L-1

with LOD of 7 × 10-12 mol L-1. The analytical performance toward MTX is

comparable to the biosensor presented in paper I with the added advantages

of shortening the time for surface modification and the stability of the grafted

layer. The reproducibility and accuracy of measurements using this sensor is

still challenging and has very much to do with how the sensor surface is

prepared.

The connection of 4-CBD concentration and CV scan rate with the appearance

of reduction peak(s) on GCE and number of monolayers of electrode grafting

was demonstrated in paper III. The number of reduction peaks, one or two

in CVs, is dependent on the extent of the deposition of monolayers (thickness

of CP layer) deposited on the electrode during grafting. The number of

monolayers can be controlled by varying the 4-CBD concentration and the

scan rate: increased as the concentration increased, but decreased when the

scan rate increased. Controlling the thickness of grafted layer on GCE can give

50

us an opportunity to optimize the analytical performance of biosensors

regarding reproducibility and accuracy.

The Bi/CP/GCE was fabricated as a new platform of electrochemical sensor

for determination of Pb2+ and Cd2+ using square wave anodic stripping

voltammetry, paper IV. The Bi/CP/GCE offered better performance in the

determination of Pb2+ and Cd2+ compared to bare GCE and GCE modified with

only CP (CP/GCE) or only Bi (Bi/GCE). The developed electrode displayed

wide linear ranges, satisfactory LODs, LOQs, repeatability and

reproducibility. This was demonstrated for the analysis of Pb2+ and Cd2+ in tap

water samples. The results obtained with the sensor were comparable with

ICP-OES making it a cheap, simple, and environmental-friendly alternative.

6. Future Perspectives

Although biosensors that use non-faradaic EIS and SVD show the possibility

of detecting MTX at very low levels further validation and improvement of its

performance are still required. Prior to validation, optimization of the

different parameters should be considered. Parameters that need to be

optimized include: 1) number of CV scans for grafting CP on the electrode, 2)

concentration of 4-CBD, lower concentrations than what normally is used (<

2.5 mM), 3) concentration of anti-MTX antibody for immobilization, 4) time

for antibody immobilization, 5) type of blocking polymer, and 6)

concentration of blocking polymer. In this case, the criteria that need to be

considered for evaluation should be the linear range obtained by SVD. The

established calibration model should be used for predicting the concentration

of test solutions with known concentrations of MTX as demonstrated in

paper II. The impact of each parameter should be compared using the

RMSEP or relative error of the predicted concentration of MTX as a measure.

The performance of the sensor should at the end be validated using real

patient samples and addressed in terms of reproducibility, selectivity, and

accuracy. For paper III, it will be strong evidence if a relation between

concentration/or scan rate and the number of monolayer can be confirmed by

AFM or XPS using glassy carbon plates.

51

Acknowledgements

First of all, I would like to express my thanks and gratitude to my supervisor,

Solomon Tesfalidet. You put a lot of effort to guide and to assist me in many

aspects over this long educational journey. When I felt down about my

experiment results and study, you always gave me encouragement,

inspiration, and motivation that kept me moving forward. I remember the

great patience and open mind you have when explained or discussed things.

This did not only open the way for expressing my opinions, but also allowed

me keep discussing with you until all my doubts have gone. Thank you so

much for taking time out of your busy schedule to revise the manuscripts and

thesis. Without your heartfelt support and help this thesis would have been

impossible.

I extremely grateful to my co-supervisor, Britta Lindholm Sethson and

William Siljebo. Britta, I have learned a lot from you about electrochemistry

out of our discussions. I appreciate your valuable guidance and comments as

well as contribution to scientific papers. Thank you, William, for your

technical help, and thoughtful discussion.

Thanks should also go to the committee for my annual evaluation, Gerhard

Gröbner and Bertil Eliasson. Thanks for your interest and suggestion for my

works. Bertil, thanks for teaching me how to use IR spectrometer and how to

interpret its spectrum.

Big thanks to Paul Geladi and Kenichi Shimizu for your contributions. Paul,

thanks for your kind explanation about multivariate data analysis and help for

correction those parts in our papers. Kenichi, I so enjoyed working and

discussing with you. The way you asked for detail have directed me to the right

track and given me new ideas from those questions. Your immense help, good

suggestion, and constructive feedback are appreciated.

I would like to thank Avni Berisha from Pristina University, Kosovo, who

initiated the idea of using diazonium, showed me how to synthesis and

characterize it electrochemically.

52

I gratefully acknowledge my Thai lecturers and friends who gave me very

helpful advice, practical suggestions, and discussion.

Thanks to Swedish International Development Cooperation Agency (SIDA)

via the International Science Program (ISP) for providing the scholarship, and

Chemistry Department, Faculty of Science and Technology, Umea University,

for admitting me to be a PhD student and good working environment.

Many thanks to the administrators and the study councilors at Chemistry

Department, Umea University, for their friendly help and facilitating in some

administration aspects. Working with them I felt very comfortable.

Special thanks to master students in my group: Besart Shatri for having done

some extra experiments of the 2nd project, and Huyền Nguyễn who worked so

hard in lab for the 4th project and also shared her viewpoint for problem-

solving. I also thank the assistance of my former and present colleagues,

Mohammad Yaser Khani Meynaq, Suman Paul, and Dong Wang. Thanks to

all friends for the fun times we had together during Fika, lunch or dinner. Your

jokes made me laugh crazily even in my tough times. My life here would be

very boring without knowing you.

Finally, and most importantly, I wish to thank my parents, sisters, and

brothers. Dad, thanks that you have never stopped pushing me toward success

and have believed I can do it. Your hope on me was one of the main reasons

that stopped me from giving up my PhD. Thanks, Mom, for your patience,

understanding and endless love. You are a wonderful person in my life.

53

References

[1] C.I.L. Justino, A.C. Duarte, T.A.P. Rocha-Santos, TrAC Trends in Analytical Chemistry.85 (2016) 36-60, DOI: 10.1016/j.trac.2016.04.004.

[2] D. Zhu, Q. Li, X. Pang, Y. Liu, X. Wang, G. Chen, Anal Bioanal Chem.408 (2016) 5593-5600, DOI: 10.1007/s00216-016-9660-3.

[3] B. Byrne, E. Stack, N. Gilmartin, R. O'Kennedy, Sensors (Basel).9 (2009)

4407-4445, DOI: 10.3390/s90604407.

[4] C.M. Mihailescu, D. Stan, R. Iosub, C. Moldovan, M. Savin, Talanta.132 (2015) 37-43, DOI: 10.1016/j.talanta.2014.08.067.

[5] E. Dock, J. Christensen, M. Olsson, E. Tonning, T. Ruzgas, J. Emneus, Talanta.65 (2005) 298-305, DOI: 10.1016/j.talanta.2004.07.002.

[6] J.J. Gooding, Electroanalysis.20 (2008) 573-582, DOI:

0.1002/elan.200704124.

[7] J. Liu, M.M. Chisti, X. Zeng, Anal Chem.89 (2017) 4013-4020, DOI: 10.1021/acs.analchem.6b04570.

[8] C. Fernández-Sánchez, C.J. McNeil, K. Rawson, TrAC Trends in Analytical Chemistry.24 (2005) 37-48, DOI: 10.1016/j.trac.2004.08.010.

[9] B.K. Kim, J. Li, J.-E. Im, K.-S. Ahn, T.S. Park, S.I. Cho, Y.-R. Kim, W.-Y.

Lee, Journal of Electroanalytical Chemistry.671 (2012) 106-111, DOI: 10.1016/j.jelechem.2012.02.022.

[10] F.C. Fernandes, A. Santos, D.C. Martins, M.S. Goes, P.R. Bueno, Biosens Bioelectron.57 (2014) 96-102, DOI: 10.1016/j.bios.2014.01.044.

[11] J. Barón-Jaimez, M.R. Joya, J. Barba-Ortega, Journal of Physics: Conference Series.466 (2013) 012023, DOI: 10.1088/1742-6596/466/1/012023.

[12] A.H. Alghamdi, Arabian Journal of Chemistry.3 (2010) 1-7, DOI:

10.1016/j.arabjc.2009.12.001.

[13] B. Lindholm-Sethson, J. Nystrom, M. Malmsten, L. Ringstad, A. Nelson, P. Geladi, Anal Bioanal Chem.398 (2010) 2341-2349, DOI: 10.1007/s00216-010-4027-7.

54

[14] C. Berggren, B. Bjarnason, G. Johansso, Electroanalysis.13 (2001) 173-180.

[15] V.M. Mirsky, M. Riepl, O.S. Wolfbeis, Biosensors and Bioelectronics.12 (1997) 977-989, DOI: https://doi.org/10.1016/S0956-5663(97)00053-5.

[16] P. Geladi, A. Nelson, B. Lindholm-Sethson, Anal Chim Acta.595 (2007)

152-159, DOI: 10.1016/j.aca.2007.01.037.

[17] S. Tesfalidet, P. Geladi, K. Shimizu, B. Lindholm-Sethson, Anal Chim Acta.914 (2016) 1-6, DOI: 10.1016/j.aca.2016.02.012.

[18] K. Krupinska, P. Geladi, I. Vikholm-Lundin, B. Lindholm-Sethson, Chemometrics and Intelligent Laboratory Systems.163 (2017) 49-54, DOI: 10.1016/j.chemolab.2016.11.008.

[19] H. Yanagisawa, R. Kurita, T. Kamata, D. Kato, O. Niwa,

Electrochemistry.82 (2014) 949-953, DOI: 10.5796/electrochemistry.82.949.

[20] B. Bansod, T. Kumar, R. Thakur, S. Rana, I. Singh, Biosens Bioelectron.94 (2017) 443-455, DOI: 10.1016/j.bios.2017.03.031.

[21] E. Arkan, R. Saber, Z. Karimi, M. Shamsipur, Anal Chim Acta.874 (2015) 66-74, DOI: 10.1016/j.aca.2015.03.022.

[22] C. Cao, Y. Zhang, C. Jiang, M. Qi, G. Liu, ACS Appl Mater Interfaces.9

(2017) 5031-5049, DOI: 10.1021/acsami.6b16108.

[23] S.-F. Zhou, X.-J. Han, Y.-Q. Liu, Journal of Alloys and Compounds.684 (2016) 1-7, DOI: 10.1016/j.jallcom.2016.05.152.

[24] O. Ouerghi, M.F. Diouani, A. Belkacem, A. Elsanousi, N. Jaffrezic-Renault, Journal of Biomaterials and Nanobiotechnology.07 (2016) 1-12, DOI: 10.4236/jbnb.2016.71001.

[25] J. Cecchetto, F.C. Carvalho, A. Santos, F.C.B. Fernandes, P.R. Bueno,

Sensors and Actuators B: Chemical.213 (2015) 150-154, DOI: 10.1016/j.snb.2015.02.068.

[26] S.K. Arya, P.R. Solanki, M. Datta, B.D. Malhotra, Biosens Bioelectron.24 (2009) 2810-2817, DOI: 10.1016/j.bios.2009.02.008.

[27] M. Braiek, K.B. Rokbani, A. Chrouda, B. Mrabet, A. Bakhrouf, A. Maaref, N. Jaffrezic-Renault, Biosensors (Basel).2 (2012) 417-426, DOI: 10.3390/bios2040417.

55

[28] D. Jiang, J. Tang, B. Liu, P. Yang, X. Shen, J. Kong, Biosensors and Bioelectronics.18 (2003) 1183-1191, DOI: 10.1016/s0956-5663(02)00253-1.

[29] B. Ozcan, B. Demirbakan, G. Yesiller, M.K. Sezginturk, Talanta.125 (2014) 7-13, DOI: 10.1016/j.talanta.2014.02.067.

[30] J. Pinson, F. Podvorica, Chem Soc Rev.34 (2005) 429-439, DOI:

10.1039/b406228k.

[31] J. Pinson, Attachment of organic layers to materials surfaces by reduction of diazonium salts, Wiley: Hoboken, NJ, USA2012.

[32] G. Liu, T. Böcking, J.J. Gooding, Electroanalytical Chemistry.600 (2007) 335-344, DOI: 10.1016/j.jelechem.2006.09.012.

[33] D.M. Shewchuk, M.T. McDermott, Langmuir.25 (2009) 4556-4563, DOI:

10.1021/la8040083.

[34] A.L. Gui, G. Liu, M. Chockalingam, G. Le Saux, J.B. Harper, J.J. Gooding, Electroanalysis.22 (2010) 1283-1289, DOI: 10.1002/elan.200900615.

[35] S. Phal, B. Lindholm-Sethson, P. Geladi, A. Shchukarev, S. Tesfalidet, Analytica Chimica Acta.987 (2017) 15-24, DOI: https://doi.org/10.1016/j.aca.2017.08.034.

[36] S. Phal, B. Shatri, A. Berisha, P. Geladi, B. Lindholm-Sethson, S.

Tesfalidet, Electroanalytical Chemistry.812 (2018) 235-243, DOI: 10.1016/j.jelechem.2017.12.072.

[37] T. Menanteau, E. Levillain, A.J. Downard, T. Breton, Phys Chem Chem Phys.17 (2015) 13137-13142, DOI: 10.1039/c5cp01401h.

[38] T. Menanteau, M. Dias, E. Levillain, A.J. Downard, T. Breton, Physical Chemistry C.120 (2016) 4423-4429, DOI: 10.1021/acs.jpcc.5b12565.

[39] B.L. Hanssen, S. Siraj, D.K.Y. Wong, Reviews in Analytical Chemistry.35

(2016) 1-28, DOI: 10.1515/revac-2015-0008.

[40] D.R. Thévenot, K. Toth, R.A. Durst, G.S. Wilson, Biosensors and Bioelectronics.16 (2001) 121-131, DOI: https://doi.org/10.1016/S0956-5663(01)00115-4.

[41] Y. Li, J. Xu, C. Sun, RSC Advances.5 (2015) 1125-1147, DOI: 10.1039/c4ra13080d.

56

[42] A. Rudnitskaya, Sensors | biomimetic sensor arrays, in: P. Worsfold, C. Poole, A. Townshend, M. Miró (Eds.) Encyclopedia of analytical science (third edition), Academic Press, Oxford, 2019, pp. 154-160.

[43] C.I.L. Justino, A.C. Freitas, R. Pereira, A.C. Duarte, T.A.P. Rocha Santos, TrAC Trends in Analytical Chemistry.68 (2015) 2-17, DOI: 10.1016/j.trac.2015.03.006.

[44] S. Mittal, H. Kaur, N. Gautam, A.K. Mantha, Biosens Bioelectron.88

(2017) 217-231, DOI: 10.1016/j.bios.2016.08.028.

[45] A. Touhami, Nanomedicine. 374-403.

[46] R. Kazemi-Darsanaki, A. Azizzadeh, M. Nourbakhsh, G. Raeisi, M. AzizollahiAliabadi, Journal of Biology and Today's world.2 (2012) 20-23, DOI: 10.15412/J.JBTW.

[47] N. R.Stradiotto, H. Yamanaka, M. Valnice B. Zanoni, J. Braz. Chem.

Soc.14 (2003) 159-173.

[48] M.C. Canbaz, C.S. Simsek, M.K. Sezginturk, Anal Chim Acta.814 (2014) 31-38, DOI: 10.1016/j.aca.2014.01.041.

[49] F. Euring, W. Russ, W. Wilke, U. Grupa, Procedia Food Science.1 (2011) 1188-1194, DOI: 10.1016/j.profoo.2011.09.177.

[50] G. Oliveira, B. Janegitz, V. Zucolotto, O. Fatibello-Filho, Open

Chemistry.11 (2013), DOI: 10.2478/s11532-013-0305-5.

[51] J.-E. Im, J.-A. Han, B.K. Kim, J.H. Han, T.S. Park, S. Hwang, S. In Cho, W.-Y. Lee, Y.-R. Kim, Surface and Coatings Technology.205 (2010) S275-S278, DOI: 10.1016/j.surfcoat.2010.08.006.

[52] Z. Zou, J. Kai, M.J. Rust, J. Han, C.H. Ahn, Sensors and Actuators A: Physical.136 (2007) 518-526, DOI: 10.1016/j.sna.2006.12.006.

[53] O.A. Farghaly, R.S. Abdel Hameed, H. Abu-Nawwas Abd-Alhakeem, Int.

J. Electrochem. Sci.9 (2014) 3287-3318.

[54] G. Hanrahan, D.G. Patil, J. Wang, J Environ Monit.6 (2004) 657-664, DOI: 10.1039/b403975k.

[55] J.S. Daniels, N. Pourmand, Electroanalysis.19 (2007) 1239-1257, DOI: 10.1002/elan.200603855.

57

[56] E.B. Bahadir, M.K. Sezginturk, Biosens Bioelectron.68 (2015) 62-71, DOI: 10.1016/j.bios.2014.12.054.

[57] H.R.S. Lima, J.S. da Silva, E.A. de Oliveira Farias, P.R.S. Teixeira, C. Eiras, L.C.C. Nunes, Biosens Bioelectron.108 (2018) 27-37, DOI: 10.1016/j.bios.2018.02.034.

[58] Z.O. Uygun, H.D. Ertuğrul Uygun, Sensors and Actuators B:

Chemical.202 (2014) 448-453, DOI: 10.1016/j.snb.2014.05.029.

[59] E.P. Randviir, C.E. Banks, Analytical Methods.5 (2013) 1098, DOI: 10.1039/c3ay26476a.

[60] Application Note: Basic of Electrochemical Impedance Spectroscopy.(2010) 1-17.

[61] S. Helali, (2013), DOI: 10.5772/53751.

[62] Electrochemical Impedance Spectroscopy (EIS): part 3 - Data Analysis.(2011) 1-3.

[63] S.M. Rezaei Niya, M. Hoorfar, Electrochimica Acta.188 (2016) 98-102, DOI: 10.1016/j.electacta.2015.11.142.

[64] J.R. MACDONALD, J. Eleciroanal. Chem.223 (1987) 25-50.

[65] G. March, T.D. Nguyen, B. Piro, Biosensors (Basel).5 (2015) 241-275, DOI: 10.3390/bios5020241.

[66] D.W.M. Arrigan, Analyst.119 (1994) 1153-1966.

[67] Application notes s-7, subject: Square wave voltammetry, Princeton

Applied Research, 801 S. Illinois Avenue, Oak Ridge, TN 37830. https://www.ameteksi.com/library/application-notes/princeton-applied-research.

[68] E. Rahmana, I. Hussain, S.K. Manirul-Haque, Der Pharma Chemica.7 (2015) 44-55.

[69] F. Harnisch, S. Freguia, Chem Asian J.7 (2012) 466-475, DOI: 10.1002/asia.201100740.

[70] A.S. Hameed, (2016) 31-45, DOI: 10.1007/978-981-10-2302-6_2.

58

[71] I.O. K'Owino, O.A. Sadik, Electroanalysis.17 (2005) 2101-2113, DOI: 10.1002/elan.200503371.

[72] P. B. Luppa, L. J. Sokoll, D. W. Chan, 314 (2001) 1-26.

[73] A. Ulman, Chemical Reviews.96 (1996) 1533-1554, DOI:

10.1021/cr9502357.

[74] A.E. Radi, X. Munoz-Berbel, V. Lates, J.L. Marty, Biosens Bioelectron.24 (2009) 1888-1892, DOI: 10.1016/j.bios.2008.09.021.

[75] L. Pilan, E.-M. Ungureanu, G. Bidan, Nonlinear Optics and Quantum Optics.39 (2009) 41-55.

[76] A.L. Gui, M. Chockalingam, J.J. Gooding, Modifying gold with 4-

(trimethylammonio)-phenyl by aryl diazonium salts via reductive deposition, Nanoscience and Nanotechnology (ICONN), 2010 International Conference on, IEEE, 2010, pp. 116-119.

[77] A. Berisha, M. Chehimi, J. Pinson, F. Podvorica, Electrode surface modification using diazonium salts, Electroanalytical chemistry, CRC Press2015, pp. 115-224.

[78] L. Janíková-Bandžuchová, R. Šelešovská, Analytical Letters.49 (2016) 122-134, DOI: 10.1080/00032719.2014.996812.

[79] L. Gao, Y. Wu, J. Liu, B. Ye, Electroanalytical Chemistry.610 (2007) 131-

136, DOI: 10.1016/j.jelechem.2007.07.030.

[80] Y. Wei, L. Luo, Y. Ding, X. Si, Y. Ning, Bioelectrochemistry.98 (2014) 70-75, DOI: 10.1016/j.bioelechem.2014.03.005.

[81] E. Gaies, N. Jebabli, Journal of Drug Metabolism & Toxicology.3 (2012), DOI: 10.4172/2157-7609.1000125.

[82] E. Asadian, S. Shahrokhian, A. Iraji Zad, F. Ghorbani-Bidkorbeh, Sensors

and Actuators B: Chemical.239 (2017) 617-627, DOI: 10.1016/j.snb.2016.08.064.

[83] S.C. Howard, J. McCormick, C.H. Pui, R.K. Buddington, R.D. Harvey, Oncologist.21 (2016) 1471-1482, DOI: 10.1634/theoncologist.2015-0164.

[84] I. Duran Meras, A. Espinosa Mansilla, M.J. Rodriguez Gomez, Anal Biochem.346 (2005) 201-209, DOI: 10.1016/j.ab.2005.07.038.

59

[85] K. Schmiegelow, Br J Haematol.146 (2009) 489-503, DOI: 10.1111/j.1365-2141.2009.07765.x.

[86] Y.D. Li, Y. Li, N.S. Liang, F. Yang, Z.P. Kuang, J. Chromatogr B Analyt Technol Biomed Life Sci.1002 (2015) 107-112, DOI: 10.1016/j.jchromb.2015.08.017.

[87] E. Begas, C. Papandreou, A. Tsakalof, D. Daliani, G. Papatsibas, E.

Asprodini, Journal of Chromatographic Science 52 (2014) 590-595, DOI: 10.1093/chromsci/bmt081.

[88] N. K. LARIYA, G.P. AGRAWAL, International Journal of Pharmacy and Pharmaceutical Sciences.7 (2014) 433-446.

[89] M. Montemurro, M.M. De Zan, J.C. Robles, Journal of Pharmaceutical Analysis.6 (2016) 103-111, DOI: 10.1016/j.jpha.2015.12.001.

[90] A. Alcantara, R. Vitor, E. Vieira, I. Martins, Braz. J. Pharm. Sci.46 (2010)

731-740, DOI: 10.1590/S1984-82502010000400015.

[91] D.A. El-Hady, N.A. El-Maali, R. Gotti, C. Bertucci, F. Mancini, V. Andrisano, Journal of Pharmaceutical and Biomedical Analysis.37 (2005) 919-925, DOI: 10.1016/j.jpba.2004.07.046.

[92] A. Yusuf, S.N. Alvi, M.M. Hammami, World Journal of Pharmacy and Pharmaceutical Sciences 3 (2014) 125-135.

[93] F. Albertioni, B. Pettersson, O. Beck, C. Rask, P. Seideman, C. Peterson,

Journal of Chromatography B: Biomedical Sciences and Applications.665 (1995) 163-170, DOI: https://doi.org/10.1016/0378-4347(94)00507-2.

[94] T. Hirai, S. Matsumoto, I. Kishi, Journal of Chromatography B.690 (1997) 267-273.

[95] H. Aboleneen, J. Simpson, D. Backes, Journal of Chromatography B: Biomedical Sciences and Applications.681 (1996) 317-322, DOI: https://doi.org/10.1016/0378-4347(95)00580-3.

[96] H. Li, W. Luo, Q. Zeng, Z. Lin, H. Luo, Y. Zhang, J Chromatogr B Analyt

Technol Biomed Life Sci.845 (2007) 164-168, DOI: 10.1016/j.jchromb.2006.07.026.

[97] E.A. McCrudden, S.E. Tett, Journal of Chromatography B: Biomedical Sciences and Applications.721 (1999) 87-92, DOI: https://doi.org/10.1016/S0378-4347(98)00439-3.

60

[98] Z. Yu, D. Westerlund, K.-S. Boos, Journal of Chromatography B: Biomedical Sciences and Applications.689 (1997) 379-386, DOI: https://doi.org/10.1016/S0378-4347(96)00357-X.

[99] J. Bluett, I. Riba-Garcia, K. Hollywood, S.M.M. Verstappen, A. Barton, R.D. Unwin, Analyst.140 ( 2015) 1981-1987, DOI: 10.1039/c4an02321h.

[100] M.A. Al-Ghobashy, S.A. Hassan, D.H. Abdelaziz, N.M. Elhosseiny, N.A.

Sabry, A.S. Attia, M.H. El-Sayed, Journal of Chromatography B.1038 (2016) 88-94, DOI: 10.1016/j.jchromb.2016.10.035.

[101] P. Guo, X. Wang, L. Liu, M.G. Belinsky, G.D. Kruh, J.M. Gallo, Journal of Pharmaceutical and Biomedical Analysis.43 (2007) 1789-1795, DOI: https://doi.org/10.1016/j.jpba.2006.12.034.

[102] G. Chen, J.P. Fawcett, M. Mikov, I.G. Tucker, Journal of Pharmaceutical and Biomedical Analysis.50 (2009) 262-266, DOI: https://doi.org/10.1016/j.jpba.2009.04.025.

[103] I. Rodin, A. Braun, A. Stavrianidi, O. Shpigun, Journal of

Chromatography B.937 (2013) 1-6, DOI: https://doi.org/10.1016/j.jchromb.2013.07.026.

[104] X. Mo, Y. Wen, B. Ren, J. Chen, C. Xu, X. Wang, M. Huang, X. Chen, Journal of Chromatography B.907 (2012) 41-48, DOI: https://doi.org/10.1016/j.jchromb.2012.08.032.

[105] E. Sonemoto, N. Kono, R. Ikeda, M. Wada, Y. Ueki, K. Nakashima, Biomed Chromatogr.26 (2012) 1297-1300, DOI: 10.1002/bmc.2700.

[106] R.C. Schofield, L.V. Ramanathan, K. Murata, M. Grace, M. Fleisher,

M.S. Pessin, D.C. Carlow, Journal of Chromatography B.1002 (2015) 169-175, DOI: 10.1016/j.jchromb.2015.08.025.

[107] R. Bouqui, G. Deslandes, B. Nieto Bernldez, C. Renaud, E. Dailly, P. Jolliet, Anal. Methods.6 (2013) 178-186, DOI: 10.1039/c3ay40815a.

[108] P. Koufopantelis, S. Georgakakou, M. Kazanis, C. Giaginis, A. Margeli, S. Papargiri, I. Panderi, Journal of Chromatography B.877 (2009) 3850-3856, DOI: https://doi.org/10.1016/j.jchromb.2009.09.034.

[109] A. Barbieri, L. Sabatini, P. Indiveri, R. Bonfiglioli, V. Lodi, F.S. Violante,

Rapid Commun Mass Spectrom.20 (2006) 1889-1893, DOI: 10.1002/rcm.2527.

61

[110] E.d. Boer, S.G. Heil, B.D. van Zelst, R.J.W. Meesters, B.C.P. Koch, M.L.t. Winkel, M.M.v.d. Heuvel-Eibrink, T.M. Luider, R.d. Jonge, Therapeutic Drug Monitoring.34 (2012) 432-439.

[111] F. Palmisano, T.R.I. Cataldi, P.G. Zambonin, Journal of Chromatography B: Biomedical Sciences and Applications.344 (1985) 249-258, DOI: https://doi.org/10.1016/S0378-4347(00)82025-3.

[112] S. Chen, Z. Zhang, D. He, Y. Hu, H. Zheng, C. He, Luminescence.22

(2007) 338-342, DOI: 10.1002/bio.968.

[113] S.G. Chakrabarti, I.A. Bernstein, Clinical Chemistry.15 (1969) 1157-1161.

[114] S.M. Sabry, M. Abdel-Hady, M. Elsayed, O.T. Fahmy, H.M. Maher, Journal of Pharmaceutical and Biomedical Analysis.32 (2003) 409-423, DOI: https://doi.org/10.1016/S0731-7085(03)00239-5.

[115] I. Duran Meras, A. Espinosa Mansilla, F. Salinas Lopez, M.J. Rodriguez

Gomez, Anal Bioanal Chem.373 (2002) 251-258, DOI: 10.1007/s00216-002-1348-1.

[116] J. Yang, X. Tan, W.-C. Shih, M.M.-C. Cheng, Biomed Microdevices 16 (2014) 673-679, DOI: 10.1007/s10544-014-9871-3).

[117] Z. Szakacs, B. Noszal, Electrophoresis.27 (2006) 3399-3409, DOI: 10.1002/elps.200600128.

[118] J. Rodríguez Flores, G.C. Peñalvo, A.E. Mansilla, M.J.R. Gómez, Journal

of Chromatography B.819 (2005) 141-147, DOI: 10.1016/j.jchromb.2005.01.039.

[119] J. Rodríguez Flores, J.J. Berzas Nevado, I. Durán Merás, M.J. Rodríguez Gómez, Journal of Separation Science.28 (2005) 658-664, DOI: 10.1002/jssc.200400059.

[120] F. Sczesny, G. Hempel, J. Boos, G. Blaschke, Journal of chromatography. B, Biomedical sciences and applications.718 (1998) 177-185.

[121] M.C. ROACH, P. GOZEL, R.N. ZARE, Journal of Chromatography:

Biomedical Application.426 (1988) 129-140.

[122] Z. Zhu, H. Wu, S. Wu, Z. Huang, Y. Zhu, L. Xi, J Chromatogr A.1283 (2013) 62-67, DOI: 10.1016/j.chroma.2013.01.085.

62

[123] M.P. Borgman, M.F. Hiemer, A.R. Molinelli, J.C. Ritchie, S.A. Jortani, Ther Drug Monit.34 (2012) 193-197.

[124] S.d. Lathouder, A.H. Gerards, E.R. de Groot, M.G. Valkhof, B.A.C. Dijkmans, L.A. Aarden, Scandinavian Journal of Rheumatology.33 (2009) 167-173, DOI: 10.1080/03009740310004865.

[125] P. Bore, R. Rahmani, J.-P. Cano, S. Just, J. Barbet, Clinica Chimica

Acta.141 (1984) 135-149, DOI: https://doi.org/10.1016/0009-8981(84)90006-8.

[126] Z. Song, Y. Wang, Y. Dong, K. Xu, H. Long, C. Deng, Y. Yin, S.A. Eremin, M. Meng, R. Xi, Analytical Methods.8 (2016) 162-170, DOI: 10.1039/c5ay02270c.

[127] F. Wang, Y. Wang, K. Lu, X. Wei, B. Ye, Journal of Electroanalytical

Chemistry.674 (2012) 83-89, DOI: 10.1016/j.jelechem.2012.04.010.

[128] F. Wang, Y. Wu, J. Liu, B. Ye, Electrochimica Acta.54 (2009) 1408-1413, DOI: 10.1016/j.electacta.2008.09.027.

[129] Y. Wang, J. Xie, L. Tao, H. Tian, S. Wang, H. Ding, Sensors and

Actuators B: Chemical.204 (2014) 360-367, DOI: 10.1016/j.snb.2014.07.099.

[130] A.A. Ensafi, F. Rezaloo, B. Rezaei, Journal of the Taiwan Institute of Chemical Engineers.78 (2017) 45-50, DOI: 10.1016/j.jtice.2017.05.031.

[131] B. Rafique, A.M. Khalid, K. Akhtar, A. Jabbar, Biosens Bioelectron.44 (2013) 21-26, DOI: 10.1016/j.bios.2012.12.028.

[132] J. Wang, P. Tuzi, M.S. Lin, T. Tapia, Talanta.33 (1986) 707-712.

[133] S. Wang, Z. Qi, H. Huang, H. Ding, Analytical Letters.45 (2012) 1658-1669, DOI: 10.1080/00032719.2012.677790.

[134] R. Šelešovská, L. Bandžuchová, T. Navrátil, Electroanalysis.23 (2011) 177-187, DOI: 10.1002/elan.201000440.

[135] Y. Guo, Y. Chen, Q. Zhao, S. Shuang, C. Dong, Electroanalysis.23 (2011)

2400-2407, DOI: 10.1002/elan.201100259.

[136] H. Ghadimi, B. Nasiri-Tabrizi, P.M. Nia, W.J. Basirun, R.M.A. Tehrani, F. Lorestani, RSC Adv.5 (2015) 99555-99565, DOI: 10.1039/c5ra18109g.

63

[137] V. Buzea, M. Florescu, M. Badea, Sensing in elecctroanalysis.7 (2012) 57-69, DOI: 978-80-7395-564-9.

[138] K. Deibler, P. Basu, Eur J Inorg Chem.2013 (2013) 1086-1096, DOI: 10.1002/ejic.201200997.

[139] J.M. Braun, R.S. Kahn, T. Froehlich, P. Auinger, B.P. Lanphear, Environ

Health Perspect.114 (2006) 1904-1909, DOI: 10.1289/ehp.9478.

[140] M. Rafati Rahimzadeh, M. Rafati Rahimzadeh, S. Kazemi, A.A. Moghadamnia, Caspian J Intern Med.8 (2017) 135-145, DOI: 10.22088/cjim.8.3.135.

[141] Guidelines for drinking-water quality [electronic resource]: Incorporating 1st and 2nd addenda, vol.1, recommendations, 3rd ed., World health organization (WHO), page: 318, 394, Geneva, 2008.

[142] S. Bouden, A. Chausse, S. Dorbes, O. El Tall, N. Bellakhal, M. Dachraoui,

C. Vautrin-Ul, Talanta.106 (2013) 414-421, DOI: 10.1016/j.talanta.2013.01.021.

[143] Y. Wu, N.B. Li, H.Q. Luo, Sensors and Actuators B: Chemical.133 (2008) 677-681, DOI: 10.1016/j.snb.2008.04.001.

[144] L. Fan, J. Chen, S. Zhu, M. Wang, G. Xu, Electrochemistry Communications.11 (2009) 1823-1825, DOI: 10.1016/j.elecom.2009.07.026.

[145] C. Innuphat, P. Chooto, ScienceAsia.43 (2017) 33, DOI:

10.2306/scienceasia1513-1874.2017.43.033.

[146] S. Lee, J. Oh, D. Kim, Y. Piao, Talanta.160 (2016) 528-536, DOI: 10.1016/j.talanta.2016.07.034.

[147] G. Zhao, H. Wang, G. Liu, RSC Advances.8 (2018) 5079-5089, DOI: 10.1039/c7ra12767g.

[148] A.M. Ashrafi, K. Vytřas, International Journal of electrochemical

Sciences.8 (2013) 2095-2103.

[149] J. Sochr, M. Machková, Ľ. Machyňák, F. Čacho, Ľ. Švorc, Acta Chimica Slovaca.9 (2016) 28-35, DOI: 10.1515/acs-2016-0006.

[150] L. Yu, P. Zhang, H. Dai, L. Chen, H. Ma, M. Lin, D. Shen, RSC Advances.7 (2017) 39611-39616, DOI: 10.1039/c7ra06269a.

64

[151] Z. Wang, G. Liu, L. Zhang, H. Wang, International Journal of Electrochemical Science.7 (2012) 12326-12339.

[152] C. Zhang, Y. Zhou, L. Tang, G. Zeng, J. Zhang, B. Peng, X. Xie, C. Lai, B. Long, J. Zhu, Nanomaterials (Basel).6 (2016), DOI: 10.3390/nano6010007.

[153] R. Ouyang, International Journal of Electrochemical Science.(2018)

1423-1440, DOI: 10.20964/2018.02.02.

[154] K.C. Honeychurch, Analytical Methods.7 (2015) 2437-2443, DOI: 10.1039/c4ay02987a.

[155] M. Cao, A. Fu, Z. Wang, J. Liu, N. Kong, X. Zong, H. Liu, J.J. Gooding, The Journal of Physical Chemistry C.118 (2014) 2650-2659, DOI: 10.1021/jp411979x.

[156] AOAC, Aoac guideline for single laboratory validation of chemical

methods for dietary supplements and botanicals, 2002, pp. 1-38.