Accepted Manuscript

Title: Simultaneous electrochemical sensing of thallium, lead

and mercury using a novel ionic liquid/graphene modified
electrode

Author: H. Bagheri A. Afkhami H. Khoshsafar M. Rezaei S.J.

Sabounchei M. Sarlakifar

PII: S0003-2670(15)00295-0
DOI: http://dx.doi.org/doi:10.1016/j.aca.2015.03.004
Reference: ACA 233786

To appear in: Analytica Chimica Acta

Received date: 6-11-2014

Revised date: 3-2-2015
Accepted date: 1-3-2015

Please cite this article as: H.Bagheri, A.Afkhami, H.Khoshsafar, M.Rezaei,

S.J.Sabounchei, M.Sarlakifar, Simultaneous electrochemical sensing of thallium, lead
and mercury using a novel ionic liquid/graphene modified electrode, Analytica Chimica
Acta http://dx.doi.org/10.1016/j.aca.2015.03.004

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Simultaneous electrochemical sensing of thallium, lead and mercury using a

novel ionic liquid/graphene modified electrode

H. Bagheri,a,* A. Afkhami,b H. Khoshsafar,b M. Rezaei,bS.J. Sabounchei,bM. Sarlakifarb

a
Chemical Injuries Research Center, Baqiyatallah University of Medical Sciences, Tehran, Iran.

b
Faculty of Chemistry, Bu-Ali Sina University, Hamedan, Iran.

* Corresponding author: Tel./Fax: +9821824825.

E-mail address: [email protected]; [email protected] (H. Bagheri).

Graphical abstract

Highlights

 A sensor using ionic liquid, graphene and phosphorus ylide was constructed.

 The surfacemorphology and properties of the electrode characterized.

 Simultaneous determination of Tl+, Pb2+ and Hg2+ in real samples was investigated.

 Itis auseful method for determining of these analytes without overlapping problem.

 In comparison to previous methods, this method has wonderful figures of merit.

1
Abstract

In the present manuscript, an electrochemical sensor for the sensitive detection of Tl+, Pb2+ and

Hg2+ is described. A new composite electrode has been fabricated using graphene, 1-n-

octylpyridinum hexafluorophosphate (OPFP), and [2,4-Cl2C6H3C(O)CHPPh3](L), as a new

synthetic phosphorus ylide. The physicochemical and electrochemical characterizations of

fabricated sensor were investigated in details. The advantages of the proposed composite

electrode are its ability in simultaneous electrochemical detection of Tl+, Pb2+ and Hg2+ with

good selectivity, stability and no need for separating of the three species from complex mixtures

prior to electrochemical measurements. The analytical performance of the proposed electrode

was examined using square wave voltammetry. Tl+, Pb2+ and Hg2+ can be determined in linear

ranges from 1.25×10 -9 to 2.00×10-7mol L-1. Low detection limits of 3.57×10-10mol L-1 for Tl+,

4.50×10 -10mol L-1for Pb2+ and 3.86×10 -10mol L-1for Hg2+ were achieved. Finally, the proposed

electrochemical sensor was applied to detect trace analyte ions in various water and soil samples

with satisfactory results.

Keywords: Electrochemical sensor; Chemically modified carbon paste electrode; Heavy metals;

Graphene ; Simultaneous determination of heavy metals

2
1. Introduction

Thallium, mercury and lead are well-known toxic contaminants in ecosystems and even their

low-levels have carcinogenic, nervous, reproductive and teratogenic toxicity to human being [1-

3]. Moreover, Tl+ toxicity is even higher than that of Hg 2+ and Pb2+ to the biosphere [2,3].

Therefore, knowledge of the quantity of these metal ions in real samples is important because

their toxicological effects depend on their concentration and chemical form [4,5]. High sensitive

spectroscopic methods allow determining metals at levels of sub-µg L-1 [6,7]. However, the

atomic spectrometric methods are expensive and not suitable for in-situ analysis of metals [6-8].

The electrochemical methods as an alternative of spectroscopic techniques have been

accepted as one of the more efficient tools to determine heavy metal ions due to their excellent

sensitivity, short analysis time, portability and low cost and maintenance [8,9]. Anodic stripping

voltammetry (ASV)at the mercury electrode is a well-established technique for the determination

of metal ions at low levels [9,10]. However, the usage of mercury is complicated in real samples

due the presence of other metal ions. Organics also are also potential interfering species [10-13].

For example, interfering metal ions in thallium analysis are including lead, cadmium, copper,

bismuth, indium, titanium and iron [14]. Therefore, various mercury-free electrodes have been

developed over the past few decades [13-15]. A large variety of modified, composite, doped and

thin film electrodes have been reviewed by Brainina et al. [8].

There are a great number of challenges for simultaneous electrochemical determination

of Tl+, Pb 2+ and Hg2+ ions in presence of other ions. One major difficulty is that voltammetric

peaks corresponding to the oxidation/reduction of these ions are considerably overlapped in

many cases. This makes the accurate measurement of peak heights complicated, so the

concentration determinations remain unsatisfactory [15,16]. For example, on gold and mercury

3
electrodes, the Tl+ signal partially overlaps those of Pb2+ and Cd2+, whereas on silver electrode,

in non-complexing medium, the Tl+ signal is only about 0.07 V more negative than that of Cd2+

[17,18]. Chemometric methods have been used to deconvolve overlapped current peaks [19,20].

They can be helpful, but unfortunately these approaches can result in low precision. On the other

hand, the reduction and adsorption competition of metal ions at the electrode surface

makesnonlinear the relationship between the response and their concentrations [15,16,19,20].

These causesthe determinations complex, time-consuming or in some cases less precise [15,16].

The performance of ASV using a chemically modified electrode (CME) is

stronglyaffected by the electrode materials [21]. Chemically modified matrices are usually

utilized to increase the surface area, conductivity and sensitivity of electrochemical sensors and

to improve selective interactions between sensing layer and analytes [22-24]. Since the first use

of carbon paste electrode (CPE) by Adams (1958) [25], it has been widely applied in the

electroanalytical community due to its low cost, ease of fabrication, flexible substrate for

modification, high sensitivity, renewable surface and etc [26]. Generally, the concept of a paste

made up with carbon material and a non-conductive and inert binder. By substituting non-

conductive organic binders (e.g. oil) with suitable conductive binder (e.g. ionic liquids), low cost

and easy-to-fabricate electrodes have been produced, with the advantages of high conductivity

and sensitivity, fast electron transfer and good antifouling ability for electroanalysis compared to

traditional oil-based electrodes. Ionic liquids (ILs) can be used as not only the pasting binders of

CPE but also the extractant of metal ions [27-29].

Also, the recent emergence of nanomaterials has opened new horizons in designing

electrochemical sensors based on CMEs [30,31]. Graphene (Gr), as a next generation electronic

material, which is also the basic structure of all graphitic materials, is a one-atom-thick planar

4
sheet of sp2 bonded carbon atoms in a honeycomb crystal lattice. Gr exhibits excellent electron

transfer promoting ability for some species and excellent catalytic behaviour toward small

bio/molecules, which makes Gr extremely attractive for electrochemical sensors [32,33]. CMEs

based on Gr take advantages of the increased electrode surface area, mass-transport rate and fast

electron transfer compared to electrodes based on bulk materials between other factors [34,35].

However, the main challenge of how to improve the selectivity of the electrochemical

sensors to Tl+, Pb2+ and Hg2+ ions remains. In order to meet this requirement, the inducing of

metal ions-specific functional groups onto the electrode with the desired properties is particularly

attractive [3,9,21,22,29]. For this purpose, L as a new synthesized ylide is introduced into the

matrix of IL/Gr paste composite (Fig. 1a) [36]. Moreover, the combination of Gr and OPFP with

L can provide excellent electrochemical platforms for target heavy metals sensing due to the

combination of enlarged active surface area, strong adsorptive capability and the specific

complexing ability of the L in analysis of environmental samples.

2. Experimental

The details of some experimental sections (2.1 to 2.6) [36-40] are given in supporting

information.

2.1. Analytical procedure

Square wave voltammetry (SWV) and cyclic voltammetry (CV) were carried out in a 20

mL electrochemical cell containing acetate buffer (0.2 mol L-1, pH 4.0) as supporting electrolyte

medium. SWASV was used for the detection under optimized conditions. Tl+, Pb 2+ and Hg2+

were deposited at the potential of -1.1 V for 90 s by the reduction of Tl+, Pb2+ and Hg2+. The

anodic stripping (reoxidation of metals to metal ions) of the electrodeposited metals was

5
performed in the potential range of -1.0 to 0.4 V. Before each measurement to ensure dissolution

of the remaining deposits on the surface of the modified electrode, pre-conditioning step was

performed at the potential of 0.9 V vs. Ag/AgCl for 45s. During the preconcentration and

cleaning steps, the solution in the electrochemical cell was stirred. The peak currents at potentials

about -0.76, -0.41 and 0.18 V vs. Ag/AgCl for Tl+, Pb2+ and Hg2+ were measured, respectively.

3. Results and discussion

3.1. Morphologies of the different electrodes

The SEM of CPE, Gr/CPE and IL/Gr/CPE were shown in Fig. 1. The surface of the CPE

was predominated by isolated and irregularly shaped graphite flakes and separated layers were

seen (Fig. 1b). Therefore, the charges could not be transferred along the vertical direction of

planes because of the block of non-conductive binder. On the surface of Gr modified CPE (Fig.

1c), homogeneous, flexible and typical crumpled and wrinkled sheets were observed, suggesting

excellent water dispersion ability. Water dispersible property of Gr possesses many fascinating

properties such as large specific surface area, numerous functional groups, strong adsorption

abilityand high carrier mobility[41]. The SEM image of IL/Gr/CPE showed more uniform

surface. As a liquid with good conductivity and high viscosity, OPFP is capable of better

dispersing the Gr in the paste than the conditional paraffin, thus could better bridge the Gr sheets

together [41,42]. Thus, the conductive performance of the IL/Gr/CPE was notably improved due

to a mixed electronic (carbon)-ionic (IL) contribution.

In FT-IR spectrum of Gr (Fig. 1e), the absence of band of C=O stretch at 1700 cm-1 is

related to reduction of the carboxyl groups (Gr-COOH) to hydroxymethyl (Gr-CH2OH) by

hydrazine as indicated by the appearance of bands at 2913 and 2828 cm-1 corresponding to the

6
C–H stretch vibrations of the methylene group [42]. The composite (Fig. 1 e(2)) displays that

except the bands of Gr, there are additional bands at 1520 and 1255 cm-1 corresponding to the

stretching vibrations of the C-N bonds in the aromatic system of pyridinium, and 895 cm-1 and
-1
the peaks at 790, 642 and 578 cm are corresponding to the stretching vibrations of P-F,

demonstrating the existence of both OPFP and Gr in the composite [40,43]. Also, The XRD

diffraction pattern of the prepared Gr is shown in Fig. 1f. It exhibits a broad diffraction peak at

2θ = 24.5◦ which matches well with the reported literature for the formation of Gr [44].

Electrochemical impedance spectroscopy (EIS) is a suitable technique for investigating of

the electrode surface dependent to charge transfer process (Fig. 2a). The equivalent circuit

compatible with the Nyquist diagram recorded for different electrodes in is depicted in Fig. 2a

(inset). On the CPE the charge-transfer resistance (Rct)value was got as 549 Ω. While on Gr/CPE

the Rct value was decreased to 325 Ω, which was due to the presence of Gr with good

conductivity in the carbon paste. The impedance spectra of the IL/Gr/CPE shows a very small

semicircle (Rct∼180 Ω), thatis remarkably lower than that of else electrodes. After addition of L

as a nonconductive agent into sensing composite, the value of Rct increased to about 240 Ω.

To exemplify the unique properties and performance of this composite is not only due to

the Gr but to the mixture of the Gr and the OPFP IL, it was necessary to compare prepared

electrode with other electrodes having the same loading of Gr but with other binders. To have a

reasonable judgment, another IL, 1-butyl-3-methyl-imidazolium hexafluorophosphate (BMIFP),

was chosen for this comparison which is shown in Fig. 2a. It was concluded the composite

IL/Gr/CPE is superior over the BMIFP/Gr/CPE in terms of having a better conductivity and

much lower background response. The difference in physical states of BMIFP and OPFP (liquid

to solid) might imply a decrease in ion mobility which would explain the lower conductivity

7
(Rct∼210 Ω). Another possibility is the difference in chemical structure where the BMIFP IL is

more hydrophilic relative to the OPFP which would increase the charging currents because of the

increase in double layer capacitance [38].

Also, electrochemical behaviours of different fabricated electrodes were recorded in

Fe(CN)63-/4- and KNO3 solution usingCV (Fig. 2b). On CPE, a pair of weak redox peaks with the

peak-to-peak separation (ΔEp) of 0.284 V was observed, indicating the sluggish electron transfer

rate at the interface. On the Gr/CPE (curve b), a pair of well-defined peaks appeared with

ΔEpvalue as 0.171 V. On the IL/Gr/CPE (curve c), the symmetric redox peaks and much higher

peak current than that on both CPE and Gr/CPE. ΔEp value as 0.068 V is obtained, which was

due to the synergic effects between IL and Gr and then the conductivity of the whole electrode

surface was further improved. After L was added to IL/Gr/CPE, obtained peaks currents were

relatively same with IL/Gr/CPE with ΔEp value as 0.08 V, indicating that the surface

conductivity and reaction reversibility of electrode were suitable by the presence of IL/Gr/L

(curve d).

3.2. Complexation investigations of L with Tl+, Pb2+ and Hg2+ ions

The complexation of several metal ions with L was studied in methanol solution. Typical

spectra were depicted in Fig. 3a-c. The ligand has an absorbance band with a λmax at 322 nm,

while each of cations does not show any absorption band in the investigated region.The results

showed that addition of some cations, e.g. Tl+, Hg2+ and Pb2+, due to the strong interactions with

L have apparent spectral changes, while the addition of some other metal ions, due to their lower

affinity to the L, did not cause any significant spectral changes. With the addition of metal ions

to ligand solution, a new peak at about 110, 140 and 130 nm for Tl+, Hg2+ and Pb2+, respectively

8
toward higher wavelengths in comparison to spectrum of free Lwere observed where neither

cation nor ligand have any absorption in this region. This is a clear indication for the formation

of complexes. Theintensity of peak has increased with ions addition, whereas the absorbance

intensity at 322 nm decreased. Also, Fig. S1 is useful to compare the absorbance changes of

theligand after the addition of a metal, e.g. Co2+, which it has low affinity to ligand. It is cleardue

to do not form a complex (Co2+ with L), the absorbancechanges of the L after addition of cobalt

ions is not significant.From the results observed for Tl+ at a mole ratio of 0.5, it can be

immediately concluded that a 2:1 complex of [L 2M] + is formed in methanol solution (Fig. 3a).

Also, Hg2+ and Pb2+ form the complexes with L as [LM]2+ (Figs. 3b and c). Therefore, the

authors decided to add L as a selective modifier for the simultaneous determination of Tl+, Pb2+

and Hg2+ to IL/Gr modified electrode.

3.3. Electrochemical behaviour of Tl+, Pb2+ and Hg2+ on the surface of various electrodes

It is well known that the simultaneous electrochemical detection of Tl+, Pb2+ and Hg2+is

problematic on most electrodes because of the proximity of their peaks, resulting in an

overlapped voltammetric response. To clarify the function of components in the electrodes

fabrication, the cyclic and stripping voltammograms were recorded with various electrodes and

are shown in Figs. 3d and e. Fig. 3d depicts CVs for the electrochemical oxidation of a mixture

of 1.20×10-7 mol L-1Tl+, Pb2+ and Hg2+on the surfaces of different prepared electrodes in a 0.2

mol L-1 acetate buffer solution of pH 4.0, recorded at a scan rate of 100 mV s−1. No obvious

peaks were obtained at an unmodified CPE after closed circuit accumulation for 90 s (purple

line). As can be seen, no anodic peak was observed within the investigated potential range on

L/CPE in the absence of analytes ions. In the presence of analytes ions using L/CPE, three clear

9
oxidation peaks corresponding to Tl+, Pb2+ and Hg2+appeared at -0.76, -0.41 and 0.18 V vs.

Ag/AgCl, respectively. Such voltammetric response has not been observed at the unmodified

electrode, showing that the response obtained at the modified electrodes is due to the presence of

modifier’s L onto electrode surface [45]. The resulted separation in three peak potentials is

sufficient enough (∼0.35 and 0.59 V) to achieve the accurate simultaneous determination Tl+,

Pb 2+ and Hg2+ in mixture samples in comparison with other studies. On the other hand, the

electrode response can be optimized for enhancing sensitivity, through surface modification or

nanostructuring. The addition of Gr into L/CPE caused only an increase in anodic peak currents

due to the increasing in the surface area of the electrode. On the IL/Gr/L/CPE, the peak currents

also increased (curve e), which was attributed to facilitate electron transfer between analyte ions

and electrode using IL. Also, in traditional oil-based CPEs, electron transfer can only take place

at the carbon/aqueous electrolyte interface. But the higher currents of IL/CPEs may also be:

changes in paste morphology, better interactions of analytes in the IL or the presence of

additional interface where transfer across the liquid/liquid interface can occur [28].

3.4. Optimization of experimental parameters

The determination of these metals in most electrolytes is difficult because of overlapping

of their voltammetric peaks and, hence, the analytical errors occur. In fact, several chemical and

mathematical approaches have been proposed to solve these overlapping peaks that include the

use of suitable electrolytes, extraction techniques, and others [19].

The interaction of modifier molecules with the metal ions is an interfacial process which

mainly depends on the nature of the functionalities which in turn depend on the nature and pH of

the bulk solution. The measurements were carried out using various buffers like acetate,

10
phosphate, borate and Briton-Robinson buffers. Among all these, acetate buffer solution with a

pH 4.0 was resulted well defined stripping peaks for analyte ions with a considerable peak to

peak separation. Results show pH dependency with the stripping current of metal ions at the

range of pH 2.0-6.5 in 0.2 mol L-1 acetate buffer solutions. From pH 2.0 to 4.0, the stripping

current gradually increased and the maximum current was observed at pH 4.0. Above pH 4.0, the

current decreased and the current responses became very noisy over pH 6.5. The noisy current

response might be caused by the formation of hydroxide precipitates of metal ions. It was

observed for all metal ions, and the pH of the preconcentration solution was selected as 4.0 for

all subsequent measurements. Also, it is noted that a moderately acidic environment is

satisfactory for the determination of Tl+, Pb 2+ and Hg2+ in real samples.

The dependence of the stripping peak currents of analytes on the accumulation potential

was studied. The peak currents increase rapidly as the accumulation potential becomes more

negative between -0.9 and -1.1 V. The experiments also demonstrated that a further negative

shift of the accumulation potential could not obviously increase the peak currents. However, it

could cause a relatively higher background current, which is mainly attributed to the hydrogen

evolution at such negative potentials. Moreover, because a more negative preconcentration

potential would increase the possibility of codeposition of interfering species, therefore -1.1 V

was chosen as preconcentration potential.

At the accumulation potentials of -1.1 V, a longer accumulation time would cause ions to

be reduced more completely. Subsequently, this would lead to higher peak currents. However,

when the accumulation time is extremely long, the reduced ions cover the electrode surface

effectively, and the peak currents do not change with the increasing accumulation time. Also, the

linear range is narrower for prolonging the accumulation time. Experiments with different

11
accumulation times ranging from 0 to 210 s were performed. Considering the sensitivity and the

efficiency, accumulation time of 90 s was used in following experiments.

3.5. Amount of L

The use of L as modifier can greatly improve the sensitivity and selectivity of

determinations, which is due to its functional sites. The peak current intensity was increased by

increasing the amount of L, because its concentration on the surface of the modified electrode

increased correspondingly. At 10% (w/w) of modifier, the largest peak currents were obtained.

However, the continuous increase of the amount of modifier caused a decrease in the peak

currents, because excessive amount of L may result in a decrease in the conductivity of the

electrodes. So the best ratio of L in IL/Gr/CPE composition was 10% (w/w).

3.6. Interference study

Cd2+ is generally considered as the major interference in the determination of thallium on

mercury and carbon electrodes by ASV since the corresponding oxidation potentials are in close

proximity to one another [15,46]. However, the oxidation process of Tl+, Pb 2+ and Hg2+in

presence of the IL/Gr/L/CPE, in acetate buffer, lead to a good improvement in peak resolution

when compared with other electrodes. As seen in Fig. 4a, stripping voltammograms registered

after making successive additions of Cd2+ in a solution containing fixed concentrations of Tl+,

Pb 2+ and Hg2+ show that their peak potentials and currents remain practically inalterable(±5%).

No interference has been found in Tl+ determination up to 1.50×10-5 mol L-1 Cd2+. However, at

1.80×10 -5 mol L-1 Cd2+, their peaks partially overlap, since the corresponding oxidation

potentials are in close proximity to one another. The interference of Cd2+ in the simultaneous

determination of Tl+, Pb 2+ and Hg2+ is actually largely overcome with the use of the proposed

12
sensor. In view of the usual high environmental abundance of Cd 2+ and other heavy metals in

comparison to Tl+, Hg2+ and Pb2+,we tried to increase the concentration of the potentially

interfering ions (Cu2+, Zn2+, K+, Na+, Ca2+, Mg2+, Al3+, Ni2+ Mn2+,Co2+, Cr3+, In3+, Ba2+, NH4+,

Cl-,NO3-,SO42-, PO43- and Fe3+) up to 3.50×10 -5 mol L-1 in order to evaluate directly applying this

method to environmental samples. Under the ±5.0% tolerated ratios, it was found that they had

no significant effect on the signals of Tl+, Pb2+ and Hg2+.

3.7. Analytical performance of IL/Gr/L/CPE

SW voltammograms with different concentrations of Tl+, Pb2+ and Hg2+ on the

IL/Gr/L/CPE were recorded and it was found that when concentrations of the three analyte ions

increase simultaneously, all three species exhibit oxidation peaks separately with a favorable

linear relation over the range from 1.25×10-9-2.00×10-7 mol L-1. The calibration equations and

correlation coefficients were y = 400.34x + 0.19 (x: mol L-1, y: µA), r = 0.9995 for Tl+, y =

305.78x + 0.129, r = 0.9994 for Pb2+ and y = 342.95x + 0.423, r = 0.9997 for Hg2+, respectively

(Fig. 4b). The limit of detection (LOD) of electrode was determined based on three times of

standard deviation of the blank signals to calibration slope (3s/m). LODs were calculated as

3.57×10 -10mol L-1 for Tl+, 4.50×10-10mol L-1for Pb2+ and 3.86×10-10 mol L-1for Hg2+,

respectively. Of course, even lower LODs could be achieved by prolonging the accumulation

time. Also, the comparison of the calibration curves of Tl+, Pb2+ and Hg2+, obtained individually

or simultaneously by SWV, showed similar linear concentration dependence and not only do not

interfere with each other but also the peak current values are exactly similar to those obtained

from their individual calibration plots (Fig. 5). It is indicated that the formation of intermetallic

complexes is unlikely. Such simultaneous detection of Tl+, Pb 2+ and Hg2+ without compromising

13
the sensitivity has not been reported in the literature. Thus, it can be concluded that the proposed

sensor was possible to detect Tl+, Pb 2+ and Hg2+ ions in a single stripping run successfully,

without encountering interferences in real samples.

The repeatability and reproducibility of the developed electrode were evaluated

byrepetitive measurements of 3.00×10-8mol L-1 Tl+, Pb 2+ and Hg2+ solution. For an electrode, the

relative standard deviation (RSD) was 3.6% for Tl+, 3.2% for Pb2+ and 2.9% for Hg2+ in five

times measurements. For five different electrodes prepared in a same batch, the RSD was 3.9%

for Tl+, 3.5% for Pb 2+ and 2.3% for Hg2+, respectively. The storage stability of the proposed

sensor was studied by detecting the SWASV responses of electrode to 3.00×10-8 mol L-1analyte

ions intermittently. The electrode lost about 4.8%, 5.6% and 4.9% Tl+, Pb2+ and Hg2+ of its

original responses after its storage at ambient conditions for 28 days. These results demonstrate

the high repeatability, reproducibility and stability of the IL/Gr/L/CPE.

3.8. Application to real samples

The analytical application of fabricated sensor was testedin tap water, river water and soil

samples.River water and soil samples collected from polluted sites of the river on the Hamedan,

where a cement plant is in operation nearby. It is interesting to check the Tl+, Pb2+ and Hg2+

levels at this place, since there is some population density together with some agricultural and

industrial activities. The anthropogenic activities may cause enrichment of Tl+, Pb 2+ and Hg2+ in

the environment, leading to abnormally high levels in natural matrices, as reported elsewhere. In

order to demonstrate the real performance of sensor,theobtained voltammograms of target

analytes in a real sample (Tap water) is shown in Fig. 6.It shows the fabricated sensor is able to

apply directly in environmental analysis with well function. Also, the species such as Zn2+, Cu2+

14
and Fe3+ existing often in tap water with relative higher concentrations than analytes, shown

without interference effects. Results obtained by this method were compared with those obtained

by graphite furnace atomic absorption spectrometry (GF-AAS) for Tl and Pb and cold vapor

atomic absorption spectrometry (CV-AAS) for Hg. It was concluded no significant difference

between the two methods. The results (Table 1) showed that the stripping method using the

IL/Gr/L/CPE produced satisfactory recovery results, with an average recovery of 99.6% for Tl+,

100.4% for Pb2+ and 101.9% for Hg 2+, respectively.

4. Conclusions

To the best of our knowledge there is no report on simultaneous direct electrochemical

determination of mixtures of Tl+, Pb2+ and Hg2+ ions in presence of other heavy metals such as

Cd2+, Cu2+ and some other species because of the close proximity of their relevant signals. In this

paper, a novel electrochemical sensor was fabricated based on integration of sensing platform

with minimal costs and the potential for in-situ monitoring, particularly for point-of-care

applications where simplicity must also be considered. Compared with other electrodes (shown

in Table 2), the proposed electrode shows excellent advantages such as wide linear ranges and

low LODs. The suggested sensor permits work at very low concentrations and at short deposition

time, renders superfluous the need to remove oxygen from the solutions and may improve the

resolution of adjacent peaks. Due to facile electron transfer reaction and high selectivity at

sensing platform, this sensor separates the voltammetric signals of the three analytes with

enhanced peak currents compared to other reported sensor and also prevents the surface fouling

effect caused by the intermetallic products. Moreover, fabricated sensorcould be a good

15
alternative to GF-AAS and CV-AAS, which for the determination of analyte species in complex

matrices, need too expensive and complex equipment. These made the sensor promising to be

used in routine analytical applications.

Acknowledgements

The authors wish to thank the Researches and Technology Council, Baqiyatallah

University of Medical Sciences, Tehran, for supports. Also, the authors acknowledge theCenter

of Excellence in Development of Environmentally Friendly Methods for Chemical Synthesis of

Bu-Ali Sina University for providing valuable financial and facility support to this work.

References:

[1] H. Bagheri, A. Afkhami, M. Saber-Tehrani, H. Khoshsafar, Preparation and characterization

of magnetic nanocomposite of Schiff base/silica/magnetite as a preconcentration phase for the

trace determination of heavy metal ions in water, food and biological samples using atomic

absorption spectrometry, Talanta 97 (2012) 87-95.

[2] G. Ja Lee, H.M. Lee, Y.R. Uhm, M.K. Lee, C.K. Rhee, Square-wave voltammetric

determination of thallium using surface modified thick-film graphite electrode with Bi

nanopowder, Electrochem. Commun. 10 (2008) 1920-1923.

[3] H. Bagheri, A. Afkhami, A. Shirzadmehr, H. Khoshsafar, A new nano-composite modified

carbon paste electrode as a high performance potentiometric sensor for nanomolar Tl(I)

determination, J. Mol. Liq. 197 (2014) 52-57.

16
[4] K. Duarte, C.I.L. Justino, A.C. Freitas, A.M.P. Gomes, A.C. Duarte, T.A.P. Rocha-Santos,

Disposable sensors for environmental monitoring of lead, cadmium and mercury, TrAC Trends

Anal. Chem. (2014), http://dx.doi.org/doi:10.1016/j.trac.2014.07.006.

[5] G. Aragay, J. Pons, A. Merkoci, Recent trends in macro-, micro-, and nanomaterial-based

tools and strategies for heavy-metal detection, Chem. Rev. 111 (2011) 3433-3458.

[6] M. Chamsaz, M.H. Arbab-Zavar, A. Darroudi, T. Salehi, Preconcentration of thallium (I) by

single drop microextraction with electrothermal atomic absorption spectroscopy detection using

dicyclohexano-18-crown-6 as extractant system, J. Hazard. Mater. 167 (2009) 597-601.

[7] C. Kokkinos, A. Economou, Tin film sensor with on-chip three-electrode configuration for

Voltammetric determination of trace Tl(I) in strong acidic media, Talanta 125 (2014) 215-220.

[8] Kh.Z. Brainina, N.A. Malakhova, N.Yu. Stojko, Stripping voltammetry in environmental and

food analysis, Fresenius J. Anal. Chem. 368 (2000) 307-325.

[9] H. Bagheri, A. Afkhami, H. Khoshsafar, M. Rezaei, A. Shirzadmehr, Simultaneous

electrochemical determination of heavy metals using a triphenylphosphine/MWCNTs composite

carbon ionic liquid electrode, Sens. Actuators B186 (2013) 451-460.

[10] J. Wang, Stripping analysis encyclopedia of electrochemistry, Willey VCH, Weinheim,

2007.

[11] D.Y. Li, J.B. Jia, J.G. Wang, Simultaneous determination of Cd(II) and Pb(II) by

differential pulse anodic stripping voltammetry based on graphite nanofibers-Nafion composite

modified bismuth film electrode, Talanta 83 (2010) 332-336.

[12] E.M. Ghoneim, Simultaneous determination of Mn(II), Cu(II) and Fe(III) as 2-(5’-bromo-

2’-pyridylazo)-5-diethylaminophenol complexes by adsorptive cathodic stripping voltammetry at

a carbon paste electrode, Talanta 82 (2010) 646-652.

17
[13] N. Lezi, C. Kokkinos, A. Economou, M.I. Prodromidis, Voltammetric determination of

trace Tl(I) at disposable screen-printed electrodes modified with bismuth precursor compounds,

Sens. Actuators B 182 (2013) 718-724.

[14] Q.T. Cai, S.B. Khoo, Determination of trace thallium after accumulation of thallium(III) at a

8-hydroxyquinoline-modified carbon paste electrode, Analyst 120 (1995) 1047-1053.

[15] N. Spano, A. Panzanelli, P.C. Piu, M.I. Pilo, G. Sanna, R. Seeber, A. Tapparo, Anodic

stripping voltammetric determination of traces and ultratraces of thallium at a graphite

microelectrode method development and application to environmental waters, Anal. Chim. Acta

553 (2005) 201-207.

[16] Y. Bonfil, M. Brand, E. Kirowa-Eisner, Characteristics of subtractive anodic stripping

voltammetry of lead, cadmium and thallium at silver-gold alloy electrodes, Electroanalysis 15

(2003) 1369-1376.

[17] J. Wang, B. Tian, Gold ultramicroelectrodes for on-site monitoring of trace lead,

Electroanalysis 5 (1993) 809-814.

[18] Y. Bonfil, M. Brand, E. Kirowa-Eisner, Determination of nanomolar concentrations of lead

and cadmium by anodic-stripping voltammetry at the silver electrode, Anal. Chim. Acta 457

(2002) 285-296.

[19] A. Herrero, M. Cruz Ortiz, Modeling the background current with partial least squares

regression and transference of the calibration models in the simultaneous determination of Tl and

Pb by stripping voltammetry, Talanta 46 (1998) 129-138.

[20] M. Cocchi, J.L. Hildago-Hildago-de-Cisneros, I. Naranjo-Rodriguez, J.M. Palacios-

Santander, R. Seeber, A. Ulrici, Multicomponent analysis of electrochemical signals in the

wavelet domain, Talanta 59 (2003) 735-749.

18
[21] A. Afkhami, H. Bagheri, H. Khoshsafar, M. Saber-Tehrani, M. Tabatabaee, A. Shirzadmehr,

Simultaneous trace-levels determination of Hg(II) and Pb(II) ions in various samples using a

modified carbon paste electrode based on multi-walled carbon nanotubes and a new synthesized

Schiff base, Anal. Chim. Acta 746 (2012) 98-106.

[22] A. Afkhami, H. Bagheri, A. Shirzadmehr, H. Khoshsafar, P. Hashemi, A potentiometric

sensor for Cd2+ based on carbon nanotube paste electrode constructed from room temperature

ionic liquid, ionophore and silica nanoparticles, Electroanalysis 24 (2012) 2176–2185.

[23] A. Walcarius, Electrocatalysis, sensors and biosensors in analytical chemistry based on

ordered mesoporous and macroporous carbon-modified electrodes, TrAC Trends Anal. Chem. 38

(2012) 79-97.

[24] H. Bagheri, A. Afkhami, Y. Panahi, H. Khoshsafar, A. Shirzadmehr, Facile stripping

voltammetric determination of haloperidol using a high performance magnetite/carbon nanotube

paste electrode in pharmaceutical and biological samples, Mater. Sci. Eng. C 37 (2014) 264-270.

[25] R. N. Adams, Carbon paste electrodes, Anal. Chem. 30 (1958) 1576-1576.

[26] P.D. Schumacher, K.A. Fitzgerald, J.O. Schenk, S.B. Clark, Preconcentration of f-elements

from aqueous solution utilizing a modified carbon paste electrode, Anal. Chem. 83 (2011) 1388-

1393.

[27] H. Dai, Y.M. Wang, X.P. Wu, L. Zhang, G.N. Chen, An electrochemiluminescent sensor for

methamphetamine hydrochloride based on multiwall carbon nanotube/ionic liquid composite

electrode, Biosens. Bioelectron. 24 (2009) 1230-1234.

[28] D.S. Silvester, Recent advances in the use of ionic liquids for electrochemical sensing,

Analyst 136 (2011) 4871-4882.

19
[29] A. Afkhami, H. Khoshsafar, H. Bagheri, T. Madrakian, Construction of a carbon ionic

liquid paste electrode based on multi-walled carbon nanotubes-synthesized Schiff base

composite for trace electrochemical detection of cadmium, Mater. Sci. Eng. C 35 (2014) 8-14.

[30] A. Walcarius, S.D. Minteer, J. Wang, Y.H. Lin, A. Merkoci, Nanomaterials for bio-

functionalized electrodes: recent trends, J. Mater. Chem. B 1 (2013) 4878-4908.

[31] C.I.L. Justino, T.A.P. Rocha-Santos, S. Cardoso, A.C. Duarte, Strategies for enhancing the

analytical performance of nanomaterial-based sensors, TrAC Trends Anal. Chem. 47 (2013) 27-

36.

[32] M.S. Artiles, C.S. Rout, T.S. Fisher, Graphene-based hybrid materials and devicesfor

biosensing, Adv. Drug Deliv. Rev. 63 (2011) 1352-1360.

[33] S. Kochmann, T. Hirsch, O.S. Wolfbeis, Graphenes in chemical sensors andbiosensors,

TrAC Trends Anal. Chem. 39 (2012) 87-113.

[34] N. Zhou, J. Li, H. Chen, C. Liao, L. Chen, A functional graphene oxide-ionic liquid

composites-gold nanoparticle sensing platform for ultrasensitive electrochemical detection of

Hg2+, Analyst 138 (2013) 1091-1097.

[35] A. Afkhami, H. Khoshsafar, H. Bagheri, T. Madrakian, Facile simultaneous

electrochemical determination of codeine and acetaminophen in pharmaceutical samples and

biological fluids by graphene-CoFe2O4 nancomposite modified carbon paste electrode, Sens.

Actuators B 203 (2014) 909-918.

[36] S.J. Sabounchei, M. Sarlakifar, S. Salehzadeh, M. Bayat, M. Pourshahbaz, H.R. Khavasi,

Structural, theoretical and multinuclear NMR study of mercury(II) and silver(I) complexes with

two new ambidentate phosphorus ylides, Polyhedron 38 (2012) 131-136.

20
[37] N. Maleki, A. Safavi, F. Tajabadi, High-performance carbon composite electrode based on

an ionic liquid as a binder, Anal. Chem. 78 (2006) 3820-3826.

[38] R.T. Kachoosangi, M.M. Musameh, I. Abu-Yousef, J.M. Yousef, S.M. Kanan, L. Xiao, S.G.

Davies, A. Russell, R.G. Compton, Carbon nanotube-ionic liquid composite sensors and

biosensors, Anal. Chem. 81 (2009) 435-442.

[39] W.S. Hummers, R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc. 80 (1958)

1339-1339.

[40] K.J. Huang, D.J. Niu, J.Y. Sun, C.H. Han, Z.W. Wu, Y.L. Li, X.Q. Xiong, Novel

electrochemical sensor based on functionalized graphene for simultaneous determination of

adenine and guanine in DNA , Colloids Surf. B: Biointerfaces 82 (2011) 543-549.

[41] C. Wu, D. Sun, Q. Li, K.B. Wu, Electrochemical sensor for toxic ractopamine and

clenbuterol based on the enhancement effect of graphene oxide, Sens. Actuators B 168 (2012)

178-184.

[42] T.H. Han, Y.K. Huang, A.T.L. Tan, V.P. Dravid, J.X. Huang, Steam etched porousgraphene

oxide network for chemical sensing, J. Am. Chem. Soc. 133 (2011) 15264-15267.

[43] A. Safavi, F. Sedaghati, H.R. Shahbaazi, E. Farjami, Facile approach to the synthesis of

carbon nanodots and their peroxidase mimetic function in azo dyes degradation, RSC Adv. 2

(2012) 7367-7370.

[44] A. Afkhami, H. Khoshsafar, H. Bagheri, T. Madrakian, Preparation of NiFe2O4/graphene

nanocomposite and its application as a modifier for the fabrication of an electrochemical sensor

for the simultaneous determination of tramadol and acetaminophen, Anal. Chim. Acta 831

(2014) 50-59.

21
[45] S.E. Harnung, M.S. Johnson, Chemistry and the Environment, Cambridge University Press,

2012.

[46] E.O. Jorge, M.M.M. Neto, M.M. Rocha, A mercury-free electrochemical sensor for the

determination of thallium(I) based on the rotating-disc bismuth film electrode, Talanta 72 (2007)

1392-1399.

[47] H.I. Dong, H. Zheng, L. Lin, B.X. Ye, Determination of thallium and cadmium on a

chemically modified electrode with Langmuir-Blodgett film of p-allylcalix[4]arene, Sens.

Actuators B 115 (2006) 303-308.

[48] L. Zou, Y.F. Zhang, H.I. Qin, B.X. Ye, Simultaneous determination of thallium and lead on

a chemically modified electrode with Langmuir-Blodgett film of a p-tert-butylcalix[4]arene

derivative, Electroanalysis 21 (2009) 2563-2568.

[49] T.H. Lu, H.Y. Yang, I.W. Sun, Square-wave anodic stripping voltammetric determination of

thallium(I) at a Nafion:mercury film modified electrode, Talanta 49 (1999) 59-68.

[50] V. A. Tarasova, Voltammetric determination of thallium(I) at a mechanically renewed bi-

graphite electrode, J. Anal. Chem. 62 (2007) 157-160.

[51] A. Bobrowski, M. Putek, J. Zarebski, Antimony film electrode prepared in situ in hydrogen

potassium tartrate in anodic stripping voltammetric trace detection of Cd(II), Pb(II), Zn(II), Tl(I),

In(III) and Cu(II), Electroanalysis 24 (2012) 1071-1078.

[52] L.M. de Carvalho, P.C. do Nascimento, A. Koschinsky, M. Bau, R.F. Stefanello, C.

Spengler, D. Bohrer, C. Josta, Simultaneous determination of cadmium, lead, copper, and

thallium in highly saline samples by anodic stripping voltammetry (ASV) using mercury-film

and bismuth-film electrodes, Electroanalysis 19 (2007) 1719-1726.

22
Figure Captions

Fig. 1. Structure of L (a), SEM images from surface of CPE (b), Gr/CPE (c) and IL/Gr/CPE (d). FT-
IR spectra for Gr (e1) and IL/Gr (e2), XRD pattern of prepared Gr (f).

Fig. 2. EIS of different electrodes in Fe(CN)63-/4- containing 0.1 mol L -1 KNO3 (a), inset shows the
equivalent circuit compatible with the Nyquist diagram for this system. (B) CVs of CPE (a), Gr/CPE
(b), (c) IL/Gr/CPE and L/IL/Gr/CPE (d) in a solution of 1.0×10-3mol L-1Fe(CN)63-/4- and 0.1 mol L-1
KNO3 at the scan rate of 100 mV/s.

Fig. 3.UV-Vis spectra for titration of L (6.8×10-5mol L-1) with Tl+ (a), Pb2+ (b) and Hg2+ (c) in
methanol (25.0 ◦C), respectively. (d) CVs of 1.20×10-7 mol L-1 Tl+, Pb2+ and Hg2+ with scan rate of
100 mV s-1 in 0.2 mol L-1 acetate buffer solution (pH 4.0) on the surface of various electrodes.
Curves (a) CPE in presence of analytes (b) L/CPE in absence of analytes (c) L/CPE, (d) Gr/L/CPE
and (e) IL/Gr/L/CPE in presence of analytes.

Fig. 4.(a) SWASV of 1.30×10-7 mol L-1 Tl+ on IL/Gr/L/CPE (a) in the absence of Cd2+, (b-e) in the
presence of Cd2+ (5.00×10-6-1.50×10-5 mol L-1) and (f) in the presence of Cd2+ (1.80×10-5 mol L-1).
(b) SW voltammograms of different concentrations of Tl+, Pb2+ and Hg2+ in 0.2 mol L-1 acetate buffer
solution (pH 4.0) using IL/Gr/L/CPE under optimum conditions. Inset shows calibration plot of the
corrected electrochemical peak currents as a function of analytes concentrations.

Fig. 5.SW voltammograms using IL/Gr/L/CPE under optimum conditions in solution containing
different concentrations of Pb2+ and Hg2+ in the presence constant concentration of Tl+ (a). Inset
shows calibration plot of the corrected electrochemical peak currents as a function of Pb2+ and Hg2+
concentrations. (b) is related to Tl+ and Hg2+ in the presence constant concentration of Pb2+ and (c) is
related to Tl+ and Pb2+ in the presence constant concentration of Hg2+, respectively.

Fig. 6.SW voltammograms using IL/Gr/L/CPE it tap water sample (a) and after addition of 1.00×10 -8
mol L-1 of analytes (b).

23
Table 1. Results forTl+, Pb 2+ and Hg2+ determination in water and soil samples obtained under the optimum conditions (N=10).

Added Found
Sample Analyte Recovery (%) AAS
(×10-9 mol L-1) (×10-9 mol L-1)
Tl+ 0.00 ND - ND b
10.00 10.09 (±0.48) 100.9 ND

Pb2+ 0.00 21.92 (±0.53) - ND

Tap water
10.00 32.17 (±0.41) 102.5 ND

Hg2+ 0.00 ND - ND
10.00 10.10 (±0.35) 101.0 10.22 (±0.56)
0.00 26.22 (±0.40) - ND
Tl+
10.00 36.10 (±0.42) 98.8 36.10 (±0.60)

0.00 185.87 (±0.28) - 186.50 (±0.35)

Pb2+
River water 10.00 196.03 (±0.20) 101.6 196.24 (±0.29)

0.00 24.90 (±0.24) - 24.71 (±0.48)

Hg2+
10.00 35.10 (±0.25) 102.0 35.00 (±0.44)
-
0.00 63.22 (±0.18) 63.04 (±0.29)
Tl+ 99.3
10.00 73.15 (±0.16) 72.68 (±0.28)
-
0.00 101.62 (±0.32) 101.80 (±0.40)
Soil Pb2+ 97.2
10.00 111.34 (±0.30) 112.19 (±0.35)
-
0.00 35.21 (±0.21) 35.32 (±0.65)
Hg2+ 102.8
10.00 45.49 (±0.20) 45.13 (±0.41)

a
Values in parentheses are S.D. based on ten replications.
b
Not detected.

24
Table 2. Comparison of some figures of merit related to the different modified electrodes for the (individual or simultaneous) determination
with priority of Tl+ accompanying Pb 2+ and Hg2+.

-1 -1
Linear range (mol L ) LOD (mol L )
Electrode Method
Tl+ Pb 2+ Hg2+ Tl+ Pb 2+ Hg2+ Refs.
Modified graphite
electrode with Bi SWASV 4.89×10-9 -4.89×10-7 N.M.a N.M. 1.47×10-10 N.M. N.M. [2]
nanopowder
Screen-printed electrode
DPASV N.R.b N.M. N.M. 4.40×10-9 N.M. N.M. [13]
modified with Bi
Graphite microelectrode SWASV 1.47×10-10-1.71×10-6 N.M. N.M. 4.89×10-11 N.M. N.M. [15]
Substractive
Silver-gold alloy 8.90×10-9-8.90×10-7 4.80×10 -9-4.80×10 -7 N.M.a 8.90×10-10 5.00×10 -10 N.M. [16]
ASV
Rotating-disc bismuth
SWASV 3.61×10-8 -1.50×10-7 N.M. N.M. 1.08×10-8 N.M. N.M. [46]
film
GCE coated with
DPASV 2.45×10-8-1.22×10-6 N.M. N.M. 4.89×10-9 N.M. N.M. [47]
Langmuir–Blodgett film
GCE coated with
Langmuir–Blodgett film SWASV 3.00×10-8-4.00×10-6 2.00×10 -7-2.00×10 -5 N.M. 2.00×10-8 8.00×10-8 N.M. [48]
Nafion/mercury film SWASV 2.45×10-10-4.89×10-7 N.M. N.M. 48.9×10-9 N.M. N.M. [49]
Bi-Graphite DPASV 4.89×10-8-4.98×10-6 N.M. N.M. 4.89×10-9 4.00×10-8 N.M. [50]
Sb-film SWASV 9.78×10-9-4.89×10-7 2.41×10 -8-4.83×10 -7 N.M. 3.47×10-10 7.24×10-9 N.M. [51]
Mercury film 0.00-2.45×10-6 0.00-1.93×10-6 N.M. 3.42×10-9 4.83×10 -10 N.M.
DPASV [52]
Bi film 0.00-9.78×10-7 0.00-9.65×10-7 N.M. 2.50×10-8 2.86×10 -10 N.M.
This
IL/Gr/L/CPE SWASV 1.25×10-9-2.00×10-7 1.25×10 -9-2.00×10 -7 1.25×10 -9-2.00×10-7 3.57× 10-10 4.50× 10-10 3.86× 10-10
work
a
N.M. (not measured); b N.R. (Not reported).

25
Fig. 1

26
Fig. 2

27
Fig. 3

28
Fig. 4

29
Fig. 5

30
Fig. 6

31