Anal. Bioanal. Electrochem., Vol. 11, No.

12, 2019, 1716-1734

Analytical &
Bioanalytical
Electrochemistry
2019 by CEE
www.abechem.com

Full Paper

Determination of Hg(II) by Thiourea Grafted Chitosan

Modified Carbon Paste Electrode: Reversibility and
Electrochemical Parameters Studies
Selma Hadnine,1,* Lilia Zighed,1 Hamed Abbassi,2 Saleh Rahmouni3 and Emna Zouaoui1

1
Department of petrochemistry and process engineering, University of 20 Août 1955-Skikda,
El-Hadaiek Road, BP: 26, 21000 Skikda, Algeria
2
High Institute of Environmental Sciences and Technologies (ISSTE), University of Carthage,
Technopole of Borj Cedria B. P,Borj Cedria, 2050 Ben arous, Tunisia
3
Department of Electrical Engineering, University of 20 Août 1955-Skikda, El-Hadaiek Road,
BP: 26, 21000 Skikda, Algeria

*Corresponding Author, Tel.: +213777449894

E-Mail: S-Hadnine@hotmail.com or meryamiya@live.fr

Received: 20 October 2019 / Received in revised form: 7 December 2019 /

Accepted: 10 December 2019 / Published online: 31 December 2019

Abstract- Mercury(II) was sensitively detected in acidic solutions using chitosan (Cts)
reinforced by grafting Thiourea (TU). In order to validate our developed method, the
characteristics of the carbon paste electrode (CPE) modified with (TU-gt-Cts) were
investigated by several electrochemical techniques, and the surface parameters were
determined. Cyclic voltammetry (CV) analysis were realized in the presence of [Fe(CN)6]3−/4−
and KCl as a redox probe and the effectives surfaces area (A) were calculated. Electrochemical
impedance spectroscopy (EIS) measurements indicated low interfacial charge transfer
resistance at the (TU-gt-Cts) modified electrode. The study of the scan rate effect denotes that
the electrochemical system presented a diffusion controlled process happened in conjunction
with a quasi-reversible electron transfer reaction. Electrochemical parameters of mercury on
the surface of (TU-gt-Cts/CPE) were determined with a chronoamperometry (CA) method as
D=2.22×10-5 cm2s-1 and k0=1.15×10-3 cms-1, for the diffusion coefficient and standard rate
constant of electron transfer respectively. The square wave anodic striping voltammetry
(SWASV) was checked as a sensitive detection method. Under the optimized chemical and
instrumental parameters, the stripping peak current was linear to the Hg(II) concentration in
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1717

the ranges of 5.0×10-9 to 1.0×10–6 molL–1 and from 1.0×10–6 to 5.0×10–6 molL–1 with a
detection limit (LOD) of 1.61×10–9 molL–1. The studied sensor was successfully applied for
Hg(II) detection in tap and seawater with adequate recoveries.

Keywords- Mercury detection, Thiourea Grafted Chitosan, Cyclic voltammetry, interfacial

parameters, reversibility

1. INTRODUCTION

Heavy metals can cause serious threats to plants, animals and human health, even in small
doses. Mercury ranks the first position in toxicity among all metals, not to mention its wide
distribution in the environment. Except the occupationally exposed population, there are three
main forms of mercury of different sources; methylmercury (MeHg), which usually occurs by
eating fish containing MeHg, mercury vapor (Hg0) which can be exposed by inhalation of
mercury (Hg0) found in dental amalgam and ethyl mercury (Hg) results from taking some
vaccines containing mercury [1]. The toxicity of mercury in humans can be explicated by its
deleterious effects on health and organs; as ability to cause kidney failure, brain damage, and
various disorders at the level of movement and cognition [2]. The mercury poisoning incident
in Iraq is perhaps the biggest proof [3].
Electrochemical techniques are useful for individual and simultaneous determination of
metal ions because of their high sensitivity, easy operation, ability of analyzing element
speciation and more importantly applicable in on-site monitoring of environmental micro-
pollutants [4,5]. One of such techniques is anodic stripping (ASV) voltammetry. It has been
recognized as the most sensitive electrochemical approach for mercury determination on
carbon electrode (CE), with limits of detection (LOD) goes about 10-7 gL-1 [6,7]. However, it
is considered a quantitative technique, and no qualitative information about electrochemical
reactions, can be inferred from the response of the (ASV). Therefore, it was necessary to resort
to cycle voltammetry (CV), on the basis of which several interfacial parameters can be found
and estimated; thermodynamics of redox processes, electron transfer kinetics and mass
transport proprieties [8,9].
Whether in adsorption or heavy metal sensing field, the chitosan has attracted a great
attention during recent decades. Due to the large amount of active amino and hydroxyl groups,
it exhibits a strong adhesion to heavy metals [10]. Cts and its derivatives have been applied as
modifier electrode for electrochemical detection of heavy metals [11,12]. Whether are synthetic
like resins based on polyvinylamine or natural like Cts, the polymers lose its ability to uptake
metals ions in acidic pH, due to protonation of amino groups [13]. Some cross-linking agents
such as glutaraldehyde (gt) [14,15], epichlorohydrin [16] and Tri-polyphosphate [17,18] have
been used to minimize its solubility and stabilize Cts in acid solutions, however, this process
would decrease the affinity of Cts for metals ions, particularly in the case of chemical reactions
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1718

that require an amine [19]. Therefore, the efforts have been focused on the chelating functional
groups as cross-linkers which can support the Cts ability to uptake metal ions [20].
According to the Irving-Williams concept about the stability of metal complexes, the
divalent metal ions like mercury(II) prefer to form complexions with ligands containing
oxygen, nitrogen and sulfur as donor atoms [21]. The grafting of sulfur and nitrogen
compounds on Cts has been the subject of many studies for the design of Cts-based chelating
resins [13,22], some of them were intended for mercury uptake or sensing [23,24,25]. Thiourea
(TU) is the most organosulfur compound used to prepare chitosan-based Schiff bases, thanks
to its known affinity for metal ions species due to their sulfur group, and double amide fonction
[19,26,27]. Moreover, it has a good execution and stability in acidic media [28]. (TU) has been
successfully grafted onto Cts using glutaraldehyde as a linking agent between amine groups of
both thiourea and Cts [20,27]. The effect of amount ratio of glutaraldehyde and thiourea in the
conjugation bath was studied and optimized by Guibal et al. [29,30]; where the characterization
by elementary analyses demonstrated that the maximum fraction of chelating groups elements
(N%=7.6 and S%=2.7) was determined from sample prepared with 3 g, 2ml and 1 g of thiourea,
glutaraldehyde and Cts respectively.
In this paper, Crosslinked Cts grafted thiourea was used as modifier of CPE for the
determination of trace amounts of Hg(II). In addition to traditional methodology followed in
heavy metal determination studies, which are usually limited to optimizing of the working
conditions, finding the limit of detection and test of analytical applicability and interferences
effect, we also discussed in this work the interface reactions including electrode reversibility,
mass transport mechanism, and the heterogeneous electron transfer reactions kinetics.

2. EXPERIMENTAL PART

2.1. Chemicals and apparatus

The chitosan was extracted from shrimp shell (Parapenaeus longirostris) following the
conventional steps described by Benhabiles et al. [31]. Using Sabnis and Block method based
on FTIR spectroscopy [32], the degree of deacetylation (DD) was estimated as 83%. The
viscosity-average molecular weight (Mv) of chitosan was calculated as 78 KDa based on
Mark–Houwink equation [ɳ] = k. (Mv)a, where the values of empirical viscometer constants
k= 0.076 and a= 0.76 were used according to the experiences of Rinaudo et al. [33].
Glutaraldehyde (gt) was supplied from sigma. Nujol oil was of analytical reagent grade and
was purchased from buck scientific. The graphite powder of 325 meshes was supplied by
Asbury Carbons. A 10-2 M Hg2+ aqueous stock solution was prepared by dissolving mercury
nitrate in reagent-grade water. Thiourea and all other chemicals were of analytical reagent
grade; purchased from Biochem chemopharma, and were used without additional purification.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1719

The modified polymer samples were characterized by FT-IR spectroscopy using a Perkin
Elmer 1600 FTIR spectrometer (400- 4000 cm-1).Voltammetric and impedance measurements
were performed with a VOLTALAB PGZ-301(Radiometer analytical), controlled by computer
using the VoltaMaster.4 software. Electrochemical cell was assembled with a conventional
three-electrode system including a carbon paste electrode (CPE), Ag/AgCl (3.0 M KCl) and
platinum electrode as the working; reference and auxiliary electrode. A WTW InoLab model
730 pH meters was also used.

2.2. Preparation of Thiourea grafted Chitosan

Based on the results of the study conducted by Guibal et al. [29,30]; three grams of Thiourea
were introduced in 20 mL of demineralized water, 2 ml of glutaraldehyde solution (50% v/v in
water) was added and the mixture was reacted for 24 h before addition of 1 g of chitosan, then
the solution was placed under agitation for 48 h. Finally, Thiourea grafted Chitosan (TU-gt-
Cts) was separated by filtration rinsed several times with demineralized water and dried at
about 50 °C overnight. Cross-linked Cts (CCts) with glutaraldehyde; was prepared as described
by Janegitz et al. [34] with little modification. Briefly, 1 g of chitosan was suspended in 20 mL
of demineralized water containing 1 mL of glutaraldehyde (50% v/v), the mixture was placed
under agitation for 12 h, and then the solid was filtered, rinsed several times with water and
dried overnight at about 40 °C.

2.3. Preparation of modified CPEs

The chemically modified electrodes were prepared by mixing TU-gt-Cts, CCts or Cts with
graphite powder and Nujol oil at a ratio of 20:65:15 (w/w) in an agate mortar. The mixture was
manually homogenized for at least 25 min. The prepared carbon paste was firmly filled into a
PVC tube. A copper wire was attached to the paste and extended outside the tube to ensure the
electrical contact. The surface of the composite electrode was rubbed with clean paper until
shining. In order to minimize the background current, several scans were applied [5]. Before
each use, the electrode surface was treated with abrasive paper and then polishing paper.

3. RESULTS AND DISCUSSION

3.1. FTIR analysis

The achievement of the "thiourea/glutaraldehyde" crosslinking as polymer base Schiff on

chitosan au levels of the free amine -NH2 is can be confirmed by comparison between the peaks
characteristics of the chitosan powder and the corresponding peaks on CCts and TU-gt Cts
spectra. The three spectra showed a broad band between 3000 Cm-1 and 3600 Cm− 1.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1720

Scheme 1. Schematic illustration of chitosan modification and (TU-gt-Cts)/GCE preparation

For Cts spectrum, the band is more intense and contains two separate small peaks at 3395
and 3360 Cm− 1, which are assigned to a typical O_H and N_H stretching vibrations on the
unmodified polymer backbone. The other two bands in the same spectral range showed a clear
lower absorption intensity with a little shift to tow new peaks at 3360 cm−1 for CCts and 3320
cm−1 for TU-gt-Cts, proving involvement of Cts amino groups in the crosslinking process with
glutaraldehyde for CCts [35] and with polymeric Schiff base for TU-gt-Cts. The band around
2875 cm-1 representing C-H groups stretching vibration is appeared in the three spectra. The
spectrum of Cts shows small peaks at around 1645 cm−1 (C=O stretch of amide I) and 1325
cm−1 (C-N stretch of amide III), characteristics of residual N-acetyl groups. The absorption at
1590 cm−1 was attributed to N-H bending vibrations of primary amines [36] on chitosan
structure. The clear band at 1645 cm−1 and 1565 Cm−1 in CCts represented the C=N of imine
group that formed between carboxyl of glutaraldehyde and amine of chitosan under what is
called Schiff base reaction. Because of a large number of C=N bonds on TU-gt-Cts, the bands
at 1629 cm− 1 and 1530 cm− 1 are appeared with increased in intensity, which can assigned to
stretching vibrations of C=N of Schiff’s base moiety and C-N of thiourea moiety [19]. An
absorption band in the general -C=S stretching region (1050–1300 cm−1) was often appeared
in spectra of the compounds in which a nitrogen atom is attached to the -C=S group [29]. This
explains the apparent increase in the peak intensity of TU-gt-Cts spectra in this region
compared to CCts. indicating the successful of thiourea grafted chitosan resin formation.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1721

100

98

96

Transmittance %
94
Cts
92 CCts
TU-gt-Cts
90

88

86

84

4000 3500 3000 2500 2000 1500 1000

Abs (Cm-1)

Fig. 1. FT-IR spectra of (Cts), (CCts) and (TU-gt-Cts)

3.2. Comparison among different modified electrodes for Hg (II) detection

To evaluate the performance of the new modifier, TU-gt-Cts/CPE was compared under
identical conditions with bare CPE, Cts/CPE and CCts/CPE, The SWASV responses of 5 µM
of Hg2+ in 0.2 M HCl at the surface of different electrodes were investigated. As can be seen
in (Fig. 2), SWASV peak current of all voltammograms appeared with peak potential between
-0.6 and -0.50 V. The modifiers promote the sensitivity of the electrode many times more than
using bare CPE.

400

350
CPE
300 Cts/CPE
CCts/CPE
250 TU-gt-Cts/CPE
current (µA)

200

150

100

50

-0,4 -0,3 -0,2 -0,1 0,0 0,1 0,2 0,3 0,4 0,5
E vs,Ag/AgCl (V)

Fig. 2. SWV profile of 5.0 µM of Hg+2 at CPE, CTs/CPE, CCts/CPE, and TU-gt-Cts/CPE.
Conditions: ʋ=50 mV.s-1, [HCl]=0.2 M, ΔE=5 mV, tacc=1 min

As regards to Cts/CPE, the modest peak height can be explained by protonation of chitosan
amine groups in acid pH of supporting electrolyte, which leads to increase of electrostatic
repulsion with Hg2+ ions [37]. The protonation was manifested on the erosion of the electrode
surface, which prevents its reuse more than one time. Grafting of chitosan deplete both -OH
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1722

and -NH2 groups [38], the retention of metal ions will be limited to free hydroxyl groups in its
crosslinked structure. The decrease in current can explicated by inhibition of ions transport by
CCts particles dispersed in the carbon paste, so that the efficiency of the latter is better alone.
A clear superiority can be seen when comparing (TU) grafted Cts with crosslinked Cts
electrodes, the peak current on TU-gt-Cts/CPE has nearly to 2 times greater than that of
CCts/CPE; the difference in the mechanism of metal attachment can explain this response
disparity: in the case of crosslinked Cts, the metal was fixed through ion-exchange reactions,
while, for TU-gt-Cts chelating resin, the attachment can results through several binding
mechanisms including adsorption, electrostatic attraction and chelation as well as ion-
exchange [30].

3.3. Electrochemical characterization of the modified electrodes in a K3[Fe(CN)6] solution

The ideal chemical and electrochemical behavior of the (ferro/ferric cyanide) system, made
it eligible to be used as a model in electrode areas determination [39]. For 1 mM of K3[Fe(CN)6]
in 0.1 M KCl as supporting electrolyte, the cyclic voltammograms of all electrodes displayed
a well-defined redox peaks; one attributed to the oxidation of Fe(CN)64−and the other to
reduction of Fe(CN)63− as showed in (Fig. 3a). Adding (TU-gt-Cts) exhibits excellent reaction
with ferro/ferric cyanide ions versus the bare of carbon paste electrode, proving the efficacy of
combination between Thiourea and Cts. While, the addition of cross-linked chitosan to the
carbon paste has resulted in a decrease of electrode performance; corresponding to the
previously recorded observation and thus enhances the proposed interpretation. The obvious
difference between CCts/CPE and Tu-gt-Cts/CPE in favor of the latter, may be attributed to
the higher affinity of sulfur atom of thiourea as a soft base towards Hg(II) ion as soft Lewis
acid; according to Pearson’s theory [40]. The peak potential spacing (ΔEp) values were
measured as 0.166 V for TU-gt-Cts/CPE and 0.332 V for CPE, which verified that the Schiff
base resin gives rise to an improvement in electron transfer rate of Fe (CN)63-/4-redox. The
quasi-reversibility of the (Fe3+/Fe2+) redox reaction was demonstrated by; the ratio value
between anodic and cathodic peak of currents density (Ipa/ Ipc), which was close to unity [41],
and the peak potentials separation (ΔEp), which was higher than (0.059/n)V.
The electroactive surface area of the constructed electrodes was calculated using Randle-
Sevcik equation (1), for a quasi-reversible process [8]:

𝐼𝑝 = 2.65 × 105 AD1/2 n3/2 Cʋ1/2 (1)

Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1723

1200
120,0µ
a 1050 b bare CPE
CCts/CPE
100,0µ
900 Tu-gt-Cts/CPE
80,0µ CPE
60,0µ CCts/CPE 750
Tu-gt-Cts/CPE
40,0µ 600
current (A)

-Z`` (ohm)
20,0µ 450
0,0
300
-20,0µ
150
-40,0µ
0
-60,0µ
-80,0µ -150

-100,0µ -300
-0,2 0,0 0,2 0,4 0,6 0 500 1000 1500 2000 2500 3000 3500 4000 4500
E vs,Ag/AgCl (V) Z` (ohm)

Fig. 3. (a) cyclic voltammograms. Scan rate: 100 mVs-1 and (b) Nyquist plots for 1 mM
Fe(CN)63-/4- in a solution containing of 0,1 M KCl at the surface of prepared electrodes; EIS
condition: frequency range: 100 kHz to 0.1 Hz; perturbation amplitude: 5 mV

Where Ip refers to the peak current (A), n is the electron involved number, A is the effective
surface area (cm2), D is the diffusion coefficient of K3[Fe(CN)6] in 0.1 molL−1 KCl solution
(6.2×10−6 cm2s−1), C is the concentration of K3[Fe(CN)6](mol cm−3) and ʋ is the scan rate (V/s).
Using the slope of Ip vs. ʋ1/2 curve. The effective surface area (A) of different electrodes
was calculated as 0.287, 0.242 and 0.337 cm2 for CPE, CCts/PE and TU-gt-Cts/CPE
respectively.
Electrochemical impedance spectroscopy (EIS) is frequently used for studying the
electrode/solution interfaces kinetic by measuring the charge transfer and mass transport
properties. Here, charge transfer resistance (Rct) values were measured employing Fe (CN)63-
/4-
as a electrochemical probe. (EIS) measurements were registered at frequencies ranging from
100 mHz to100 KHz with amplitude of 5 mV in a solution of 0.1 M KCl containing 1 mM of
K3Fe (CN)6. (Fig. 3b) depicts the Nyquist plot of various electrodes, where, Zi and Zr are the
imaginary and real impedance of the electrochemical cell. At high frequencies, the semicircular
shape from Nyquist plots is resulted from the limiting transfer of electrons, and its diameter is
proportional to the charge transfer resistance (Rct), while the straight line part at low
frequencies, which is known as Warburg impedance, is related to the limited process of
diffusion or adsorption mass transport, between the electrode surface and the solution
containing the redox probe [42,43]. The Rct value measured at CPE was 1.5 KΩ, modification
of carbon paste with CCts increased the resistance of electrons transport (Rct=3.05 KΩ); As
most functional groups of CCts structure are occupied due to crosslinking, the electric
conductivity of the electrode surface has become limited. While, the diameter of semicircle
dwindled dramatically in the case of (TU-gt-Cts) as modifier, where the value of (Rct) has
become 170 Ω (curve c), which can be explained by the great affinity of the target ion to Schiff
base at the electrode surface, that positively reflected on the process of electron transport.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1724

3.4. Electrochemical properties and reversibility studies

3.4.1. Effect of scan rate on the cyclic voltammetric behavior

Valuable information including electrochemical mechanism generally can be obtained

from the relationship between peak current and scan rate (v). The effect of scan rate study on
the anodic (Ipa) and cathodic (Ipc) peak current was executed using CV, taking 102 µM Hg(II)
in 0.2 M HCl in the range of 10 to 150 mVs-1, as shown in (Fig. 4a).

Fig. 4. a) CV responses of TU-gt-Cts/CPE in 0.2 M HCl at scan rates (inner to outer) 10, 20,
35, 50, 75, 100, 150 mVs-1; Inset: (b) the plot of anodic and cathodic peak current vs. square
root of scan rate; (c) the plot of logarithm of anodic and cathodic peak current (A) vs. logarithm
of scan rate (V/s); (d) variation of anodic and cathodic potential (V) vs. logarithm of scan rate
(V/s), and (e) Tafel plots derived from the rising part of the voltammograms recorded at the
scan rates of 20, 35, 75 and 100 mVs-1

It can be seen that the redox peak currents showed a linear relationship with square root of
the potential sweep rate (Fig. 4b), demonstrating a diffusion-controlled process. The linear
regression equations were obtained as:

Ipa (µA)=94.02 ʋ1/2 (mVs-1) – 99.56 (R2=0.995) and

Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1725

Ipc (µA =-63.90 ʋ1/2 (mVs-1) –3.13 (R2=0.998).

Diffusion controlled reaction can also confirmed from the log Ip(A) versus log(scan rate)
(V/s) plot (Fig. 4). This logarithmic plots gave the slopes=0.66 with R2=0.979, and 0.50 with
R2=0.994 for anodic and cathodic current, The values of slopes are very close to 0.5, which is
the theoretical value of the ideal diffusion controlled reaction [44]. On the other hand, the peak-
to-peak separation (ΔEp) was lightly increased with elevating scan rates, the ΔEp increased
from 63 mV to 169.5 mV with increasing scan rate from 0.010 Vs-1 to 0.150 Vs-1. Where; the
anodic peaks advanced progressively towards more positive potentials and the cathodic peaks
retreated progressively to more negative potentials (Fig. 4d). The ratio Ipc/Ipa values of the
reverse-to-forward peak current is less than 1, indicating slow heterogeneous electron transfer
processes [9]. Therefore, these results reveal that the electrochemical process is quasi-
reversible.
Based on the cycle voltammograms at different scan rate, the anodic transfer coefficient α
value can also be determined from the data of rising part of each cyclic, known as Tafel region,
which is influenced by electron transfer kinetics between metal ions (Hg+2) and TU-gt-Cts/CPE
[45]. (Fig. 4e) shows the Tafel plot of the anodic ascending portion from the curves data of 20,
35, 75 and 100 mV/s scan rate. This procedure uses a plot of log I(A) versus potential E(V),
and the equation (2), to determine the transfer coefficient α [46]:

𝑏 = 2.3𝑅𝑇 / (𝛼 − 1) 𝑛𝐹 (2)

The average slope of Tafel plots gives an mean of transfer coefficient value equal to α=0.55.

3.4.2. Chronoamperometry

In the case where electrode reactions are controlled by diffusion, the current can be given
by the Cottrell equation (3), and the diffusion coefficient can be calculated by the experimental
Cottrell slope of chronoamperometric I vs. t–1/2 plot [47]:

𝐼𝑝 = 𝑛𝐹𝐴𝐶𝐷1/2 𝜋 −1/2 𝑡 −1/2 (3)

where D is the diffusion coefficient (cm2s−1), A is the surface area of the electrode and C the
bulk concentration of Hg(II) (molcm−3), and the other symbols have their usual meanings.
Measurements of Hg(II) at TU-gt-Cts/CPE were carried out by setting the sensor potential at -
420 mV for three different concentrations of mercury ions in HCl 0.2 M. From the resulting
slope showed in (Fig. 5) and using Cottrell equation the mean value of the D was found to be
2.22×10−5 cm2/s.
The quasi-reversible process previously proven is usually attributed to the comparable
rates of mass transport and charge transfer; Were the mass transport coefficient, mT is
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1726

determined by equation (4) [8]. While the heterogeneous electron transfer rate constant, ko, can
be calculated by a Nicholson’s technique developed in 1965, using equation (5) [48]

𝑚𝑇 = [𝐷/(𝑅𝑇/ 𝐹ʋ)]1/2 (4)

𝛹 = 𝐾°[𝜋𝐷𝑛 𝜐𝐹 / (𝑅𝑇)]−1/2 (5)

Ψ in (eq.5) is a kinetic parameter and its values can be extracted from the Nicholson’s
working graph. As the application of this method is limited to a peak separation range of 60–
200 mV, Lavagnini et al have proposed a new approach that can make up for this limitation
and allows for the calculation of standard rate constant k° whatever be the values of (ΔEp x n)
obtained from CV curves. By combining the Nicholson's method and the Klingler-Kochi
treatment designed for irreversible systems, a new formula was given for each approach; with
different ranges of validation.

Fig. 5. Chronoamperograms obtained at TU-gt-Cts/CPE in the presence of 6.0×10-5 M,

1.3×10-4 M; and 2.5×10-4 M of Hg(II) in 0.2 M HCl. Inset: Cottrell's plot for the data from the
chronoamperograms

Formula (6) is the equivalent empirical equation of the Nicholson’s working curve and it
was used to calculate large values of Ψ, while the small ones were computed by equation (7)
which is the reorganized form of Klingler and Kochi relationship [49]:

𝛹 = (−0.6288 + 0.0021𝑋) / (1 − 0.017𝑋) (6)

𝛹 = 2.18 [𝛼/𝜋]1/2 𝑒𝑥𝑝 [−(𝛼 2 𝐹/𝑅𝑇) 𝑛 𝛥𝐸𝑝] (7)
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1727

where X in (eq.6) indicated (ΔEp x n) expressed in mV. And α in (eq.7) is the transfer coefficient
for the electrode process.
Varying the scan rate from 10 to 150 mVs-1, the ψ values were calculated from the equation
(6) or (7) each one in their ranges of validity. Using the equation (5) an average value of
heterogeneous electron transfer rate constant was found to be k0=1.15×10-3cm·s-1, which
confirms the quasi-reversibility of electrochemical process, because (10−1>k°>10−5 cms-1) [50].
And on the other hand, this value can be compared to the average value of mass transport
coefficient, were mT =3.86.10-3 cms-1.

3.5. SWASV determination of Hg(II)

3.5.1. Optimization of SWASV analysis conditions

The principal parameters that affecting the peak current value were optimized using 7.0
µM of Hg(II). The influence of the HCl concentration was studied in the range from 0.05 to
0.5 molL−1 (Fig. 6a). In this study, the best result was recorded for 0.3 molL−1.

Fig. 6. Influence of the experimental variables (HCl concentration, potential and time of
electro-deposition, step potential and pulse height, respectively) involved in the SWV method,
using the TU-gt-Cts/CPE in the presence of 7.0 µmolL−1 Hg(II)
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1728

As it is very close to the result obtained with 0.2 M, this latter was selected as the
concentration of supporting electrolyte for further experiments, to avoid excessive use of
concentrated acid.
The electro-accumulation potential effect on the anodic peak current was evaluated in the
range from 0.1 V to -1.0 V. (Fig. 6b) shows that the accumulation becomes fewer efficient
when the applied potential is near to the (Hg+2/Hg0) potential, because the potentials are not
negative sufficient at this level to electrochemical reduction of Hg+2 to Hg0; the application of
Ered=0.1 V barely gave a signal. The negative shift more then −1.0 V in electro-reduction
potential, can decreases the stripping currents and deteriorates the baseline of mercury response
due to the reduction of other chemicals [51]. Based on the (Fig. 6b) results, the potential of -
0.8 V was chosen as sufficient. The influence of electro-reduction time was also examined (Fig.
6c). The electrode response remained proportional to the time in the range from 30 s to 300 s,
with a clear decrease in the slope recorded from the third minute, indicating the beginning of
electrode surface saturation. As a result, 180 s were selected as sufficient time for Hg+2 electro-
accumulations.
Alongside the peak heights, both signal-to-noise ratio and good signal shapes must be
consider in optimization of instrumental square-wave voltammetric parameters. In the range
from 30 mV/s to 500 mV/s, the peak current intensity was linearly proportional with potential
scan rates. However, the peak started to widen and the signal took a curvy shape starting from
100 mV/s. Therefore, this value was chosen for future experiments. Concerning other
conditions, the selected optimal values have been 2 mV for step potential (interval of study 1
to 15 mV) and 45 mV as pulse height (range of investigation 10 to 50 mV).

3.5.2. Calibration Curve and Detection Limit

SWASV was employed to study the linear response and the detection limit of Hg (II) at
TU-g-Cts/CPE. The effect of Hg (II) concentration on the peak current was investigated under
the optimum conditions mentioned above. A series of stripping voltammograms of 5.0 nM to
5.0 µM Hg (II) are shown in (Fig. 7). It can be seen that the calibration plots of peak current
vs. concentration show two linear dynamic ranges from 5.0×10-9 to 1.0×10–6 molL–1 and from
1.0×10–6 to 5.0×10–6 mol L–1 (inset Fig. 7). The linear regression equations for these two regions
are:

Ipa(μA)=352.55 [Hg(II)] (µM)+10.31 (R2=0.998) and

Ipc (μA)=147.64 [Hg(II)] (µM)+204.72 (R2=0.998).

The limit of detection value was calculated based on the formula LOD=3×SD/m, where SD
is the standard deviation for the blank solution (replicate determinations=10) and m is the slope
of the analytical curve [52].
 Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1729

1500
400 1000
+2
1400 350 Ip (µA) = 352,55 [Hg ] +10,6 +2
Ip (µA) = 147,64 [Hg ]+ 204,72
2 900

1300
300 R = 0,998
c 800
2
R = 0,998
d
a

current (µA)
250

current (µA)
700
1200 200
150 600

1100 100 500

1000
50 400 blank
0 300 0,005µM
900
current (µA)

0,0 0,2 0,4 0,6

[Hg+2] (µM)
0,8 1,0
1 2 3 4 5 0,01µM
[Hg+2] (µM)
800 0,05µM
1000 0,1µM
700 b
800 0,3µM
600 0,5µM
current (µA)

600
0,7µM
500 400
1,0µM
400 200
2,0µM
300
0 3,0µM
0 1 2 3 4 5 4,0µM
200 [Hg+2] (µM)
5,0µM
100
0

-0,4 -0,3 -0,2 -0,1 0,0 0,1 0,2 0,3 0,4

E vs,Ag/AgCl (V)

Fig. 7. (a) Square wave voltammograms for various concentrations of Hg(II). Inset:
Corresponding linear calibration curves of peak current vs. Hg(II) concentration; (b) from 5nM
to 5 µM, (c) from 5.0 nM to 1.0 µM and (d) from 1.0 µM to 5.0 µM. Optimized conditions:
ʋ=100 mVs-1, [HCl]=0.2 M, ΔE=2 mV, tacc=3 min

Table 1. Comparison of TU-gt-Cts/CPE efficiency with various electrochemical modified

electrods for voltammetric determination of Hg(II) in literature

Sensor Method Linear range Detection limit Refs.

( (nM) ( nM)
Cts–SH/ MWCNTs/GCE SWASV 10 to 140 3 [15]
OCM/CPE DPASV 99.7 to 698 67.8 [53]
DNA functionalized Graphene /GCE DPVs 8 to 100 5 [54]
TFSF/ GCE DPASV 10 to 100 4 [55]
SWCNT–PhSH/Au E SWASV 5 to 90 3 [56]
BTC/ CPE SWV 50 to 200 39.9 [57]
DTTPSG/ CPE SWASV 50 to 5000 9.97 [58]
PEDOT/GO DPSV 10 to 3000 2.78 [59]
DPSG/ CPE ASV 10 to 100 7.97 [60]
TU-g-Cts/CPE SWASV 5 to 1000 and 1.61 this work
1 1000 to 5000
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1730

The Hg(II) detection limit was 1.61 nM, this LOD is lower than the reported values
obtained from several electrochemical methods using electrodes modified by chitosan
derivatives (Table 1). The TU-gt-Cts/CPE shows much lower determination limit and wider
linear range.

3.6. Validation Methods

3.6.1. Reproducibility and repeatability

To examine the stability of response efficiency versus possible changes in signal shape
resulting from differences in electrodes preparation or surface interfacial interactions over time,
the studies of electrode reproducibility and repeatability were necessary. Using the optimum
conditions previously described, the relative standard deviation (RSD) does not exceed 0.8%
for six electrodes constructed from the same carbon paste and tested to determination of 0.1µM
of Hg(II). While in the case of six electrodes prepared from different paste homogenized with
same manner, the RSD was slightly larger 2.2%. For six successive stripping voltammograms
with the same TU-gt-Cts/CPE, the RSD in peaks current was 2.1%. The long-term stability
was also studied using one modified electrode; (RSD) was estimated as 4.6% for eight
repetitive measurements over a week time interval for 2 months. The results indicate the good
quality of the surface electrode and the long stability of TU-gt-Cts modifier.

3.6.2. Effect of Interferences

The ability of our sensor to discriminate between mercury and the potential interfering
species commonly present in similar physiological environment was also examined to test the
selectivity. The percentage of current change was used to express the interference effect.
Whatever the concentration of monovalent ions such as K+, Na+, Cl-, NO3-, no evident
interference can be found for the determinations of Hg2+ at 10-6 M. The same remark has been
observed for some bivalent cations including, Cd2+, Cu2+, Co2+, Zn2+ and Ni2+, as long as the
[Mn2+]/[Hg2+] ratio is less than 10.

Table 2. Influence of interfering on the peak current response of 10–6 M Hg2+

Interfering ions Mn±1/2 Interference

[Mn ±1/2
]≤10 M
-6
[Mn±1/2]/[Hg]=10 [Mn±1/2]/[Hg]=20
k+, Na+, Cl-, NO3- No interference No interference Negligible effect

Cd2+, Cu2+, Zn2+, Co2+, Ag1+ No interference Negligible effect +4.7%

Pb2+ Negligible effect -6.4% -12.3%

Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1731

Above this ratio the effect of competition between these ions for occupying the binding
sites began to appear. Until [Mn+2]/[Hg+2]=20, the voltammetric signal change did not exceed
the tolerance limit estimated by ±5% and defined as the maximum relative error in the
analytical signal, that can be caused by the presence of foreign species. These results can justify
by several reasons; the higher affinity of schiff base as electrode surface modifier toward Hg(II)
in comparison with other ions, the difference between the diffusion coefficient of Hg(II) and
the other ions coefficients. Only lead that interferes by reducing the Hg(II) signal. This means
that there is a convergence in the complexation kinetics between these two cations, leading to
a competition in the accumulation on the electrode surface.

3.6.3. Analytical Applications

To evaluate the practical application of TU-gt-Cts/CP electrode for analyzing mercury ions
with real water samples, the laboratory tap water and seawater were used as examples for real
sample analysis. Without prior treatment, HCl (1 M) was added to the samples to give a
concentration of 0.2 M. Subsequently, mercury was spiked into the samples to prepare analysis
solutions by injection of different Hg+2 concentrations. Under the optimum conditions; resolved
current peak of 1 µM as spiked value was observed around 0.15 V for tap water and 0.18 V for
seawater. Using the analytical equation of calibration curve, the findings of spiked samples
were given in Table (3). The recoveries of mercury ions were varied in the range of 96% to
107%. These satisfactory results demonstrated the suitability of TU-gt-Cts developed matrix
for precise identification of Hg(II) ions, at trace levels in environmental monitoring of tap water
and seawater.

Table 3. Hg(II) recovery results in the samples of tap water and seawater based on the
presented sensor

Sample Spiked value/M Determined Recovery

value/M (%)
0.050 0.052±0.006 104%
Tap water
0.100 0.096±0.005 96%

0.500 0.512±0.052 102.4%

Sea water 0.100 0.107±0.0057 107%

0.500 0.532±0.011 106.4%

1.000 1.061±0.032 106.1%

Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1732

4. CONCLUSION

In this work, a sensor based on Thiourea grafted Chitosan was developed, and successfully
applied to the detection of Hg(II) in different water samples, with no matrix interference. The
electrochemical behavior of Hg(II) on the modified electrode surface was carefully examined
with determination of different electrochemical parameters. Quasi-reversible electron transfer
reaction was happened in parallel with diffusion controlled process. The influence of the
experimental variables which involved in the SWV determination of Hg(II) was also optimized.
The present sensor shows simple and low cost preparation with high sensitivity and long
stability. Additionally the TU-gt-Cts/CPE presented a very low limit of detection (1.61 nM),
that helps monitor the Hg(II) levels in real water.

Acknowledgement

The authors gratefully acknowledge the support of this work by University of 20 Août 1955
Skikda- Algeria.

REFERENCES

[1] V. Branco, S. Caito, M. Farina, J. Teixeira da Rocha, M. Aschner, and C. Carvalho, J.

Toxicol. Environ. Health 20 (2017) 119.
[2] H. H. Harris, I. J. Pickering, and G. N. George. Science 301 (2003) 1203.
[3] Amin-Zaki, L., S. Elhassani, M. A. Majeed, T. W. Clarkson, R. A. Doherty, and M.
Greenwood, Am. Acad. Pediatr 54 (1974) 587.
[4] S. Anandhakumar, J. Mathiyarasu, and K. L. N. Phani, Anal. Methods 4 (2012) 2486.
[5] M. M. Abdel-Galeil, M. M. Ghoneim, H. S. El-Desoky, T. Hattori, and A. Matsuda, J.
Chem. Biochem. 2 (2014) 25.
[6] V. Jovanovski, N. I. Hrastnik, and S. B. Hočevar, Electrochem. Commun. 57 (2015) 1.
[7] V. Jovanovski, and N. I. Hrastnik, Microchem. J. 146 (2019) 895.
[8] D. A. C. Brownson, and C. E. Banks, The Handbook of Graphene Electrochemistry,
Verlag London Ltd: Springer (2014).
[9] A. J. Bard, and L. R. Faulkner, Electrochemical methods; Fundamentals and
Applications, 2nd. ed. NewYork: John Wiley & Sons, Inc (2001).
[10] X. W. Liu, Q. Y. Hu, Z. Fang, X. J. Zhang, and B. B. Zhang, Langmuir 25 (2009) 3.
[11] K. Elmorsy, H. N. A. Hassan, I. H. I. Habib, and R. Metelka, Int. J. Electrochem. Sci. 5
(2010) 158.
[12] L. Zhenzhen, Y. Shanli, Y. Qiong, L. Shenglian, L. Chengbin, and T. Yanhong.
Microchim. Acta 180 (2013) 555.
[13] M. Weltrowski, B. Martel, and M. Morcellet, J. Appl. Polym. Sci. 59 (1996) 647.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1733

[14] J. Lopez-Cervantes, D. I. Sanchez-Machado, R. G. Sanchez-Duarte, and M. A. Correa-

Murrieta, Adsorpt. Sci. Technol. 36 (2018) 215.
[15] W. Deng, Y. Tan, Y. Li, Y. Wen, Z. Su, Z. Huang, S. Huang, Y. Meng, Q. Xie, Y. Luo,
and S. Yao, Microchim. Acta 169 (2010) 367.
[16] B. C. Janegitz, L. C. S. Figueiredo-Filho, L. H. Marcolino-Junior, S. P. N. Souza, E. R.
Pereira-Filho, and O. Fatibello-Filho, J. Electroanaly. Chem. 660 (2011) 209.
[17] W. S. Wan Ngah, and S. Fatinathan, J. Environ. Manage. 91 (2010) 958.
[18] S. T. Lee, F. L. Mi, Y. J. Shen, and S. S. Shyu, Polymer 42 (2001) 1879.
[19] L. Wang, R. Xing, S. Liu, Y. Qin, K. Li, H. Yu, R. Li, and P. Li, Carbohydr, Polymer 8
(2010) 305.
[20] A. M. Donia, A. A. Atia, and K. Z. Elwakeel, J. Hazard. Mater. 151 (2008) 372.
[21] H. Sigel, and D. B. McCormick, Acc. Chem. Res. 3 (1970) 201
[22] Y. Baba, H. Noma, R. Nakayama, and Y. Matsushita, Anal. Sci. 18 (2002) 359.
[23] J. D. Merrifield, W. G. Davids, J. D. MacRae, and A. Amirbahman, Waterr Res 38
(2004) 3132.
[24] K. Chauhan, P. Singh, and R. K. Singhal, ACS Appl. Mater. Interf. 7 (2015) 26069.
[25] M. Monier, Int. J. Biolog. Macromol. 50 (2012) 773.
[26] L. Zhou, J. Liu, and Z. Liu J. Hazard. Mater. 172 (2009) 439.
[27] M. Zhang, Y. Zhang, and R. Helleur, Chem. Eng. J. 264 (2015) 56.
[28] W. C. Schiessl, N. K. Summa, C. F. Weber, S. Gubo, C. Dücker-Benfer, R. Puchta, and
R. van Eldik, Anorg. Allg. Chem. 631 (2005) 2812.
[29] E. Guibal, T. Vincent, and R. Navarro Mendoza, J. Appl. Polym. Sci. 75 (2000) 119.
[30] E. Guibal, N. Von Offenberg Sweeney, T. Vincenta, and J. M. Tobin, React. Funct.
Polym. 50 (2002) 149.
[31] M. S. Benhabiles, R. Salah, H. Lounici, N. Drouiche, M. F. A. Goosen, and N. Mameri,
Food Hydrocolloids 29 (2012) 48.
[32] S. Sabnis, and L. H. Block, Polym. Bull. 39 (1997) 67.
[33] M. Rinaudo, M. Milas, and P. L. Dung, Int. J. Biol. Macromol. 15 (1993) 281.
[34] B. C. Janegitz, L. H. Marcolino-Junior, S. P. Campana-Filho, R. C. Faria, and O.
Fatibello-Filho, Sens. Actuators B 142 (2009) 260.
[35] Z. A. Sutirman, M. M. Sanagi, K. J. Abd Karim, and W. A. Wan Ibrahim, Carbohydrate
Polymers 151 (2016) 1091.
[36] J. B. Lambert, Introduction to organic spectroscopy, New York: Macmillan (1987).
[37] R. S. Vieira, and M. M. Beppu, Colloids Surf. A 279 (2006) 196.
[38] F. C. Vicentini, T. A. Silva, A. Pellatieri, B. C. Janegitz, O. Fatibello-Filho, and R. C.
Faria, Microchem. J. 116 (2014). 191.
[39] J. J. Van Benschoten, J. Y. Lewis, W. R. Heineman, D. A. Roston, and P. T. Kissinger,
J. Chem. Educ. 60 (1983) 772.
Anal. Bioanal. Electrochem., Vol. 11, No. 12, 2019, 1716-1734 1734

[40] A. Afkhami, H. Bagheri, H. Khoshsafar, M. Saber-Tehrani, M. Tabatabaee, and A.

Shirzadmehr, Anal. Chim. Acta 746 (2012) 98.
[41] K. Zinoubi, H. Majdoub, H. Barhoumi, S. Boufi, and N. Jaffrezic-Renault, J. Electroanal.
Chem. 799 (2017) 70.
[42] S. Shrestha, R. J. Mascarenhas, O. J. D’Souza, A. K. Satpati, Z. Mekhalif, A. Dhason,
and P. Martis, J. Electroanal. Chem. 778 (2016) 32.
[43] N. Mohammadi, M. Najafi, and N. B. Adeh. Sens. Actuators B 243 (2017) 838.
[44] H. Muhammad, I. Ahmad Tahiri, M. Muhammad, Z. Masood, M. Ali Versiani, O. Khaliq,
M. Latif, and M. Hanif, J. Electroanal. Chem. 775 (2016) 157.
[45] N. Soltani, N. Tavakkoli, N. Ahmadi, and F. Davar, C. R. Chim. 18 (2015) 438.
[46] M. M. Ardakani, H. Beitollahi, M. K. Amini, B. F. Mirjalili, and F. Mirkhalaf, J.
Electroanal. Chem. 65 (2011) 243.
[47] M. Uçar, A. O. Solak, and N. Menek, Anal. Sci.18 (2002) 997.
[48] R. S. Nicholson, J. Electroanal. Chem. 37 (1965) 1351.
[49] I. Lavagnini, R. Antiochia, and F. Magno. Electroanalysis 16 (2004) 505.
[50] J. Wang, Anal. Electrochem. 3thed, Wiley Online Library (2006).
[51] J. M. Pinilla, L. Hernhdez, and A. J. Conesa, Anal. Chim. Acta 319 (1996) 25.
[52] M. Rizk, M. M. Abou El-Alamin, H. A. M. Hendawy, and M. I. Moawad,
Electroanalysis 28 (2015) 770.
[53] I. K. Tonle, E. Ngameni, and A. Walcarius, Sens. Actuators B 110 (2005) 195.
[54] Y. Zhang, H. Zhao, Z. Wu, Y. Xue, X. Zhang, Y. He, and Z. Yuan. Biosens. Bioelectron.
48 (2013) 180.
[55] I. Cesarino, and É. T. G. Cavalheiro, Electroanalysis 20 (2008) 2301.
[56] J. Wei, D. Yang, H. Chen, Y. Gao, and H. Li, Sens. Actuators B 190 (2014) 968.
[57] M. Colilla, M. A. Mendiola, J. R. Procopio, and M. T. Sevilla, Electroanalysis 17 (2005)
933.
[58] N. L. D. Filho, and D. Ribeiro do Carmo, Electroanalysis 17 (2005) 1540.
[59] Y. Zuo, J. Xu, X. Zhu, X. Duan, L. Lu, Y. Gao, and Y. Yu, Synth. Met. 220 (2016) 14.
[60] M. Javanbakht, H. Khoshsafar, M. Ganjali, A. Badiei, P. Norouzi, and A. Hasheminasab,
Curr. Anal. Chem. 5 (2009) 35.

Copyright © 2019 by CEE (Center of Excellence in Electrochemistry)

ANALYTICAL & BIOANALYTICAL ELECTROCHEMISTRY (http://www.abechem.com)

Reproduction is permitted for noncommercial purposes.