Accepted Manuscript

Development of a novel electrochemical sensor based on

electropolymerized molecularly imprinted polymer for selective
detection of sodium lauryl sulfate in environmental waters and
cosmetic products

Soukaina Motia, Ioan Albert Tudor, L. Madalina Popescu, Roxana

Mioara Piticescu, Benachir Bouchikhi, Nezha El Bari

PII: S1572-6657(18)30244-3
DOI: doi:10.1016/j.jelechem.2018.03.069
Reference: JEAC 3985
To appear in: Journal of Electroanalytical Chemistry
Received date: 16 November 2017
Revised date: 27 March 2018
Accepted date: 29 March 2018

Please cite this article as: Soukaina Motia, Ioan Albert Tudor, L. Madalina Popescu,
Roxana Mioara Piticescu, Benachir Bouchikhi, Nezha El Bari , Development of a novel
electrochemical sensor based on electropolymerized molecularly imprinted polymer for
selective detection of sodium lauryl sulfate in environmental waters and cosmetic
products. The address for the corresponding author was captured as affiliation for all
authors. Please check if appropriate. Jeac(2017), doi:10.1016/j.jelechem.2018.03.069

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.
 ACCEPTED MANUSCRIPT

Development of a novel electrochemical sensor based on electropolymerized molecularly

imprinted polymer for selective detection of sodium lauryl sulfate in environmental

waters and cosmetic products

Soukaina Motia a,b, Ioan Albert Tudor c, L. Madalina Popescu c, Roxana Mioara Piticescu c,

Benachir Bouchikhi a, Nezha El Bari b,*

PT
a
Sensor Electronic & Instrumentation Group, Department of Physics, Faculty of Sciences,

Moulay Ismaïl University, B.P. 11201, Zitoune, Meknes, Morocco

RI
b
Biotechnology Agroalimentary and Biomedical Analysis Group, Department of Biology,

SC
Faculty of Sciences, Moulay Ismaïl University, B.P. 11201, Zitoune, Meknes, Morocco
c
Nanostructurated Materials Laboratory, National R&D Institute for Non-Ferrous and Rare
NU
Metals, Pantelimon, Ilfov, Romania
MA
E D
PT
CE
AC

*
Corresponding author: Prof. Nezha El Bari
Postal address : Faculty of sciences, Department of Biology,
B.P. 11201, Zitoune, 50003 Meknes, Morocco
Tel: +212 535 53 88 70; Fax: +212 5 35 53 68 08
Email: n_elbari@hotmail.com
 ACCEPTED MANUSCRIPT

ABSTRACT

In this study, a novel molecularly imprinted polymer (MIP) sensor for the electrochemical
detection of sodium lauryl sulfate (SLS) is reported. The MIP based screen-printed gold
electrode (Au-SPE) was prepared by an electrochemical polymerization of 2-
aminothiophenol (2-ATP) in the presence of SLS as target molecule. For control, a non-
imprinted polymer (NIP) sensor was also constructed identically by omitting the SLS
template. Moreover, the selectivity of the proposed MIP sensor was investigated by using

PT
analogues such as Ethylenediaminetetraacetic acid (EDTA), Tween 80 and Urea, revealing
satisfactory selectivity towards SLS. The electrochemical characterization of the elaborated

RI
sensor was performed using cyclic voltammetry (CV), differential pulse voltammetry
(DPV), and electrochemical impedance spectroscopy (EIS) techniques. The morphology of

SC
the MIP sensor was studied by means of Atomic Force Microscopy (AFM) and Fourier
Transform Infrared Spectroscopy (FTIR) methods. Some experimental parameters such as
NU
number of cycle for the electropolymerization, incubation time of 2-ATP, elution and SLS
incubation time, were optimized to improve the sensor performances. Under optimal
MA

conditions, the electrochemical sensor exhibits a logarithmic working range from 0.1 to
1000 pg/mL, and a detection limit of 0.18 pg/mL. Additionally, compared with reported
works, the sensor possesses remarkable properties, such as a higher sensitivity and
D

selectivity, a good reproducibility, a wider logarithmic range, a lower detection limit and a
E

long-term stability. As real application, the developed MIP sensor was well employed to
PT

determine SLS contents in environmental waters and cosmetic products samples. A UV-Vis
spectrophotometer was used as reference method. A partial least squares (PLS) technique
CE

was well employed to study the correlation between the spectrophotometer and the MIP
sensor technology with a satisfactory regression coefficient (R=0.99). The obtained results
AC

during the sensor development and applications suggested an interesting and valuable tool
for environmental analysis.

Keywords: Sodium lauryl sulfate; Molecularly imprinted polymer; Electropolymerization;

Environmental waters; Cosmetic products; UV-Vis spectrophotometer.
 ACCEPTED MANUSCRIPT

1. Introduction

Surfactants have become a very important group of compounds used in industries of modern
life. These substances possess chemical structure with two distinct regions consisting of
hydrophilic and hydrophobic parts. According to the charge of the hydrophilic group,
surfactants can be divided into four types: anionic, cationic, amphoteric and non-ionic. The
use of surfactants in different day life products is gradually increasing [1-3]. Indeed, in
manufacturing of several cosmetic and hygienic preparations, they are considered a key

PT
chemical and are therefore, widely found in household detergents, personal care products,
emulsification, lubrication and plastics industry [4]. Although they are useful and greatly

RI
needed, they may be dangerous for human health because of their chemical properties and
increasing presence in environmental waters [5].

SC
Among anionic surfactants, the sodium lauryl sulfate (SLS) has been used for about 40 years
[6]. A SLS molecule has 12 carbon atoms linked with a sulfate group, giving to the molecule
NU
amphiphilic properties. Generally, SLS is used as a foaming agent or cleanser in commonly
used cosmetic products: shower gels, shampoos, soaps, toothpastes, creams and lotions [7].
MA

Indeed, the indicated SLS concentration is remarked to vary by product and manufacturer but
typically ranges from 0.01 to 50 % and 1 to 30 % in cosmetic and cleaning products,
respectively [8]. It is a highly effective surfactant used for the removal of oily stains and
D

residues. However, it is a skin and eyes irritant [9, 10]. Some researchers have studied skin
E

irritations caused by SLS in order to determine the concentration range in which it becomes
PT

toxic. They show that application of products containing 0.5 – 10 % of SLS can cause slight
to moderate skin irritation [11, 12]. Over this range, SLS can cause skin corrosion and severe
CE

irritation. Its wide use releases it into the environment via household wastewater systems
involving an ecological risk for aquatic organisms and human beings [13]. In the
AC

environment, it is degraded through microbial activity. SLS molecule possesses a linear long
chain sulfate ester structure, which can be degraded by several bacterial strains through
desulfation and further assimilation [14, 15]. Its degradation in the environment can constitute
a toxicity against various organisms [16]. In such cases, there is a great and important need to
elaborate tools capable of detecting and determining SLS in complex samples.
To date, various methods have been proposed concerning SLS detection such as,
chromatography [17], methylene blue method [18], ion-selective electrodes [19, 20],
colorimetric and spectrometric techniques [21, 22]. Nevertheless, some of these instrumental
analyses remain usually expensive with exhaustive procedures, and require highly trained
 ACCEPTED MANUSCRIPT

technical personnel. Accordingly, to overcome these limitations, it is of great challenge to

develop rapid, inexpensive and sensitive methods for the detection of SLS in various samples.
As first alternative, electrochemical technique has advantages of easy operation, rapid
response and high sensitivity [23, 24]. In the aim of improving the selectivity, molecularly
imprinted polymer (MIP) technology is usually combined with electrochemical technique for
constructing sensors [25]. Recently, because of their high selectivity, simplicity, reliability
and sensitivity, MIPs have become an area of interest for many searchers [26-29]. The
application of MIPs in electrochemistry is rather recent and is directed to combine their

PT
intrinsic properties to selected electrochemical reactions. By this way, the polymer film will
be able to rebind a specific analyte having similar size, shape and functional groups than

RI
created sites through the matrix [30, 31]. Thus, after the molecularly imprinted membrane

SC
synthesis, different methods of its immobilization can be employed. One of them is the
electropolymerization technique. This latter is of great simplicity, uniformity and allows a
NU
controlled film thickness and deposition [32]. It can be set, as needed, by changing
electrochemical parameters [33-35]. Elsewhere, 2-aminothiophenol (2-ATP) has been utilised
to prepare MIP based sensors [36]. It can interact with many target molecules through
MA

hydrogen bond and can be electropolymerized [37, 38]. In addition, screen printed electrodes
(SPEs) have been pointed as potential sensor substrates to develop systems of disposable
D

MIPs. Some reports have been described for directly synthesizing the MIP onto the SPE via
surface imprinting method [39].
E

To the best of our knowledge, there are very few reports using electrochemical MIP sensor for
PT

SLS detection. Only two works have been proposed for SLS detection by using MIP
technology [40, 41]. For this purpose, the main goal of this work is to develop a very sensitive
CE

and selective electrochemical MIP sensor for SLS determination basing on

electropolymerization technology on Au-SPE. The MIP was designed onto the Au-SPE, and
AC

SLS molecules filled the constructed cavities by binding sites of the molecularly imprinted
film. Then, this is the first reported electrochemical MIP sensor through electropolymerization
of 2-ATP onto Au-SPE for SLS detection. The molecular imprinting sensor was not only
convenient by exhibiting a high sensitivity and selectivity toward SLS, but also allowed good
reproducibility and long term-stability. Moreover, some parameters influencing the
performances of the sensor have been carefully optimized and the obtained highly sensitive
and selective MIP sensor has been applied for SLS determination in wastewaters, river waters
and cosmetic products samples. Regarding way of SLS content determination, the correlation
 ACCEPTED MANUSCRIPT

between the sensor technique and a UV-Vis spectrophotometer has been studied by using a
statistical method named PLS.

2. Materials and methods

2.1. Reagents and solutions

Sodium lauryl sulfate (SLS), 2-Aminothiophenol (2-ATP), ethanol, phosphate buffered saline
(PBS 0.01 M), potassium ferricyanide K3[Fe (CN)6], potassium ferrocyanide K4[Fe (CN)6],
Tween 80 and Urea were purchased from Sigma Aldrich. Ethylenediaminetetraacetic acid

PT
(EDTA) was obtained from Fluka. Distilled water was used for the preparation of all solutions
and for washing. The stock solution of SLS was prepared by dissolving 1 mg of SLS in 1 mL

RI
of water. Less concentrated SLS solutions were prepared by diluting the stock solution with

SC
water. For electrochemical characterization, a standard stock solution of K3[Fe (CN)6])/ K4[Fe
(CN)6] (10 mM) was prepared using PBS (pH = 7.2).
NU
2.2. Synthesis of MIP and NIP materials onto the gold electrode

Before starting any deposit, the gold surface of the electrode was cleaned by rinsing with
MA

ethanol and water. A monolayer of 2-ATP was firstly formed [42] after dropping onto the
working gold electrodes 10 µL of a solution containing 5.4 mg of 2-ATP and 1 mL of ethanol
D

for an incubation of 12 h at 4 °C. Then, the electrodes were rinsed with ethanol and double
distilled water to remove the 2-ATP molecules that were loosely bounded. Through an
E

electropolymerization technique, both MIP and NIP films were constructed onto the modified
PT

Au-SPE by using cyclic voltammetry (CV). In order to obtain imprinted film, the
electropolymerization of MIP was carried out by using 30 µL of a solution containing 2-ATP
CE

3-/4-
(0.1 mg/mL) and SLS (1 mg/mL) in a supporting electrolyte (10 mM [Fe (CN)6] in PBS
pH 7.2). The electropolymerization step was performed in a potential range between -0.35 V
AC

and 0.80 V, at a scan rate of 100 mV/s, for 10 cycles. Moreover, to facilitate the binding and
recognition of the SLS molecules, an extraction stage of the template from the MIP film was
made by immersing the well-prepared electrode in water solution for 20 min at room
temperature. Similarly, a control non-imprinted film (Au-SPE/NIP) was prepared following
the same procedure, but without adding the template to the solution of the
electropolymerization process. This was done in order to confirm that the observed effects
during the MIP detection were only due to the imprinting features.

2.3. Spectroscopic measurements

 ACCEPTED MANUSCRIPT

To study the surface morphology and topography of the modified Au-SPE, AFM and FTIR
techniques were employed to investigate surface changes involved after the
electropolymerization and extraction steps. The AFM study was done using an apparatus
(NANOVEA, USA) having a maximum resolution of 110 μm in a scanning zone of (76
μm/76 μm) [43]. In the case of FTIR method, the study was realized at room temperature via
a FTIR spectrophotometer (HATR, ABB, MB3000 (64 scans)). The spectra were obtained in
a wavenumber range of 500–4000 cm-1 with a resolution of 4 cm-1.

PT
2.4. UV-Vis spectrophotometry analysis

As a reference method, a UV–Vis spectrophotometer was employed to measure SLS

RI
absorbance of the studied samples in order to determine their SLS contents. The

SC
environmental waters and cosmetic products samples were analysed using P Selecta
instruments UV 2005 spectrophotometer. Prior to measurements, a calibration curve was
NU
established by plotting obtained absorbance at 654 nm versus corresponding SLS
concentrations. To make measurement in real samples, 50 mL of each environmental water
sample is placed in a beaker. Then, 5 mL of chloroform is added to the solution, and the
MA

obtained mixture is stirred for 1 min with a stirrer [44]. Elsewhere, 100 mg of each cosmetic
product were dissolved in 100 mL of water and then filtered. The obtained solution was
poured into a beaker followed by the addition of PBS (pH = 7.2) and chloroform [45]. A
D

blank solution was inserted in the spectrophotometer before each absorbance measurement for
E

calibration. Afterward, by using the previous equation of the calibration curve, absorbance
PT

value of each real sample has allowed to determine their SLS contents.
CE

2.5. Electrochemical measurements

A portable PalmSens3 (Spain) interfaced with a computer was employed to make all
AC

electrochemical measurements. Fig. 1 illustrates an overall view of the system used for these
experimentations. The screen-printed gold electrodes (Au-SPE) consisted of a three-
electrode system (purchased from Dropsens), with a gold working electrode (0.19 cm2), a
saturated silver reference electrode and a gold strip plate counter electrode (0.54 cm2). Three
electrochemical techniques of characterization such as cyclic voltammetry (CV), differential
pulse voltammetry (DPV) and electrochemical impedance spectroscopy (EIS) were used
during the elaboration and detection stages to evaluate the sensor response by measuring
the current and resistance changes. DPV measurements were accomplished by scanning the
potential from -0.1 V to 0.2 V with a scan rate of 50 mV/s. The EIS analysis was measured
 ACCEPTED MANUSCRIPT

at a bias potential of 0.175 V over a frequency range of 0.1 Hz to 50 kHz and under a number
of frequencies of 50. To get the interesting informations about the Au-SPE surface
modifications, the impedance data were well fitted using the adequate Randles equivalent
circuit.

2.6. Samples analysis of the environmental waters and cosmetic products

For the real detection, the prepared MIP sensor was expected to determine SLS concentration
in wastewaters, river waters and cosmetic products samples. The environmental waters

PT
samples were collected from two wastewater discharges located in Khemisset and Rabat and
from four river waters located in Fes and Sidi kacem. They were collected in polystyrene

RI
bottles and directly used to avoid their degradation and consequent change in their intrinsic
properties. Beside, shower gel, toothpaste and shampoos were purchased from a supermarket.

SC
From each studied cosmetic products, 0.5 g was accurately weighed and dissolved in 10 mL
of methanol. After a filtration step, the supernatant was diluted in PBS (pH = 7.2) and ready
NU
for the detection stage. The pre-treated samples were incubated onto the well prepared MIP-
sensor for 30 min and the electrochemical measurement was carried out using the Palmsens
MA

(potentiostat).

3. Results and discussion

D

3.1. Mechanism of the molecular imprinting

E

The electrochemical characteristics of the stepwise MIP sensor fabrication was studied in 10
PT

mM of [Fe(CN)6]3-/4- solution containing PBS (pH = 7.2). [Fe(CN)6]3-/4- redox probe was
chosen as a marker to investigate the changes onto the electrode after each step of the sensor
CE

assembly. The preparation procedure of the MIP sensor is illustrated in Fig. 2. The whole
process can be summarized in three steps: the modification of Au-SPE with 2-ATP layer, the
AC

electropolymerization of a 2-ATP solution in the presence of SLS template onto the modified
Au-SPE, and the removal of the template molecules from the obtained MIP film. In the first
step, to improve the adhesion and the rate of the polymer formation, a monolayer of 2-ATP is
formed onto the surface of gold electrodes via Au-S bonds between gold and thiol groups (–
SH) of 2-ATP. In the second step, the MIP film is constructed onto the electrode via an
electropolymerization of a mixture solution containing 2-ATP and SLS. While the polymer
grows, SLS molecules migrate toward the working electrode and get entrapped in the polymer
matrix. Finally, they are removed from the constructed MIP, creating sites that can
specifically recognize and bind SLS. The electrochemical detection of SLS by the MIP sensor
 ACCEPTED MANUSCRIPT

was achieved by placing 10 μL of different SLS concentrations of the adopted range onto the
working electrode for 30 min. Thereafter, electrochemical measurements were carried out in
the presence of 30 μL of [Fe(CN)6]3-/4- solution as redox probe.

3.2. Optimization of experimental parameters

Several parameters influencing the performances of the MIP sensor, such as number of cycle
for the electropolymerization, incubation time of 2-ATP deposit, elution and SLS incubation
time have been optimized in order to obtain the best sensor proprieties.

PT
The change of the current peaks relative to the 2-ATP deposit times onto the gold electrode
was investigated. The 2-ATP solution was initially deposited overnight onto the Au-SPE as

RI
incubation time. This duration were optimized by using 10, 11, 12, 13, 14, 15, 16, 17 and 18
h. The experimental results in Fig. 3A showed that the optimal 2-ATP deposit time is obtained

SC
for 12 h.
The number of potential scan cycle is also an important factor that determined the thickness
NU
and morphology of the resulting MIP film. For this purpose, this parameter has also been
studied. As shown in Fig. 3B, after 10 CV cycles, the maximum current peak was achieved
MA

therefore revealing a better MIP film preparation by repeating the potential scan for 10 cycles.
A pivotal step in obtaining recognition cavities is the appropriate extraction of the template
molecule from the MIP film. Then, the extraction time was optimized, by washing thoroughly
D

the electrodes with water for different times. An increase of the current value was observed
E

from 10 to 20 min. After 20 min, the response remained stable (Fig. 3C) probably due to the
PT

fully removal of SLS molecules. According to these observations, an extraction time of 20

min was optimally considered for the removal of SLS from the polymer matrix.
CE

Moreover, the incubation time is defined as the equilibrium time needed for the uptake of SLS
in the cavities. After 20 min of extraction time, the electrodes were left in contact with a
AC

solution of SLS for different incubation times. As shown in Fig. 3D, the current peak
increased gradually up to 30 min and then remained constant due to the surface insulation.
Thus, 30 min was selected as the SLS incubation time for further experiments.

3.3. Electropolymerization of SLS imprinted film

An electrochemical polymerization process was performed during the second step of the
sensor fabrication. The recorded cyclic voltammograms during the electropolymerization in
the presence of 2-ATP and SLS are shown in Fig. 4. This was carried out in an electrolyte
solution of 10 mM [Fe(CN)6]3-/4- in PBS (pH = 7.2) within a potential range from -0.35 to 0.8
V at a scan rate of 100 mV/s for 10 cycles. The MIP film formation was monitored through
 ACCEPTED MANUSCRIPT

the changes in the current per cycle. The electrochemical polymerization process consists of
intermolecular reactions between the aniline moieties of 2-ATP (–NH2) and sulfate group of
SLS. The current gradually decreases while increasing the number of cycles, suggesting
stepwise growth of a compact polymeric film that covered the surface of the gold electrode.
This MIP formation tends to hinder the charge transfer from the electrolyte to the electrode.
In order to confirm the SLS recognition ability of the MIP, a NIP modified Au-SPE was
prepared under identical conditions but without SLS addition.
CV was used for further characterization in 5 mM [Fe (CN)6]3-/4- solution containing PBS.

PT
The obtained results are shown in Fig. 5A. As can be seen, the CV signal of the MIP electrode
is higher than those of the NIP, indicating that SLS was successfully entrapped onto the 2-

RI
ATP matrix, hindering [Fe (CN)6]3-/4- to diffuse through polymer towards Au-SPE surface.

SC
Indeed, the less important current of the NIP electrode is probably due to the absence of SLS
molecules in the polymer matrix.
NU
Impedance spectroscopy was also used to characterize the MIP and NIP modified electrodes.
Fig. 5B illustrates the Nyquist plots of the MIP and NIP electrodes. The impedance
measurements are in good concordance with the CV results as the diameter of semicircles for
MA

the Nyquist diagrams correlate with current peak variations. Nyquist representation consists
of a semicircle portion corresponding to electron-transfer and a small linear portion related to
D

the diffusion process.

E

3.4. AFM and FTIR characterization

PT

In the aim of studying the morphology and topography of the MIP sensor during its
preparation process, two instrumental techniques were utilised such as AFM and FTIR;
CE

presented AFM images in Fig. 6 show the side view of the Au-SPE surface (A), two-
dimensional (B) and three-dimensional (C) displaying the surface morphologies of the MIP
AC

film. The obtained images confirm the formation of the polymer films onto the electrodes and
offer important informations about the topography and surface roughness of the films. Indeed,
a roughness of 3 μm was observed corresponding to the electropolymerized film. The AFM
images of the extraction step show a different surface organization with a good coverage of
the electrode surface but retain a homogeneous appearance. Thus, when SLS molecules are
removed from the imprinted film, the roughness decreased to a minimum value of 2 µm (Fig.
7). According to these results, we can conclude that SLS molecules were fully extracted from
the MIP matrix.
 ACCEPTED MANUSCRIPT

Fig. 8 shows the FTIR spectroscopy of the MIP sensor development (after the
electropolymerization and extraction stages). Here, the presence of SLS molecules is
demonstrated by C-H bands (3032 and 2920 cm-1) and sulfate vibration bands S = O (993 and
1121 cm-1). The presence and absence on the spectra (A) and (B) of the sulfate functions
belonging to SLS confirm the caption of the SLS molecules and their extraction from the MIP
film. By using FTIR method, the obtained microscopic characterization results confirm that
the SLS molecules were successfully bound and then completely removed from the MIP
matrix.

PT
3.5. Molecular recognition by MIP and NIP sensors

RI
In order to check the uptake process for variable concentrations of SLS, DPV and EIS
techniques were employed.

SC
Fig. 9A shows DPV responses of the MIP-modified electrode for SLS detection in a
concentration range from 0.1 to 1000 pg/mL. During the procedure, [Fe (CN)6]3-/4- was used
NU
as the mediator between the imprinted electrode and the substrate solutions. As seen, the
redox current peaks decrease with the increase of SLS concentrations, which can be attributed
MA

to the gradual dense film covering the surface of the Au-SPE, confirming the binding of the
SLS molecules that hinders the electron transport of [Fe (CN)6]3-/4- redox probe.
In Fig. 9B, the Nyquist plots corresponding to the different SLS concentrations are illustrated.
D

They give informations about interfacial charge-transfer impedances (Rct) which were
E

determined from the diameter of the semi-circle. It was found that the Rct of the electrode
PT

increased with increasing SLS concentrations. The resistance increase of the polymer film in
the presence of SLS, an electro-inactive molecule [41] with the polymer, confirms its caption.
CE

Upon the analysis of NIP in the same conditions, the current peaks significantly decrease and
slightly vary. This observation is due to the absence of SLS during the electropolymerization
AC

involving no specific cavities creation (Fig. 10A).

Beside, Nyquist plots show simply linear plot with no semi-circular pattern and a negligible
change of the resistance values (Fig. 10B). This means that the NIP sensor was unspecific for
SLS molecules and failed to recombine them, demonstrating that responses obtained from the
MIP detection stage were only due to SLS cavities presence.

3.6. Calibration curve and detection limit

The obtained optimal conditions and parameters were exploited to fabricate the better SLS
MIP sensor in terms of performances. Therefore, the MIP sensor responses were used to plot
calibration curves.
 ACCEPTED MANUSCRIPT

For DPV, a good logarithmic relation between the MIP sensor responses and the logarithmic
of SLS concentrations in the range from 0.1 to 1000 pg/mL was obtained with a correlation
coefficient of 0.99 (Fig. 11A). In the same way, a correlation coefficient of 0.99 was obtained
for EIS technique according to Fig. 11B where Rct versus Log (concentration of SLS) is
represented. The limits of detection (LOD) and quantification (LOQ) were calculated using
the ks/m relation [46], where k = 3 for LOD and 10 for LOQ. Values of LOD and LOQ were
found to be 0.18 pg/mL and 0.61 pg/mL respectively, which indicate the good sensitivity of
the proposed method. Indeed, the elaborated MIP sensor shows better analytical performances

PT
in terms of sensitivity compared to previously reported methods [41, 47-49], as shown in
Table 1.

RI
3.7. Selectivity, reproducibility and stability of the MIP sensor

SC
To investigate the selectivity of the electrochemical MIP sensor for SLS determination, the
interferences of some similar molecular structures such as, Tween 80, urea and EDTA were
NU
considered [41]. As shown in Fig. 12, the observed current responses of the MIP sensor
toward SLS were much higher than the other analogues, indicating a better adsorption and
MA

binding capacity for the template molecules. These results certainly mean that the fabricated
MIP sensor could successfully avoid the interferences making it more suitable for specific
detection of SLS.
D

Moreover, the selectivity of this relevant sensor and the reported work [41] toward SLS
E

relative to analogues such as EDTA, tween 80 and urea, is investigated. For this purpose, their
PT

(𝐼𝑆𝐿𝑆 − 𝐼𝐴𝑛𝑎𝑙𝑜𝑔𝑢𝑒 )
relative responses ⁄𝐼 is calculated. As remarked, the selectivity of this
𝑆𝐿𝑆
CE

reported MIP work toward SLS relative to EDTA, tween 80 and urea is around 0.66 fold,
whereas in our work 0.64, 0.64 and 0.74 fold were obtained, respectively. These results
AC

demonstrate that our MIP sensor is more selective toward SLS relative to urea. Elsewhere, a
quasi-similar selectivity was found for EDTA and tween 80. In another SLS MIP sensor [40],
the selectivity performance has been slightly described.
Moreover, the reproducibility of the molecularly imprinted sensor was investigated by using a
SLS concentration of 0.1 pg/mL. Six electrodes, constructed under the same conditions, were
used to determine the corresponding current peak. Indeed, the obtained results with DPV
technique show a relative standard deviation (RSD) value of 4.9 % that confirms the
reproducibility of the sensor.
 ACCEPTED MANUSCRIPT

Stability is also an interesting parameter for MIP sensors development. In our case, the
stability of the electrode was investigated by measuring at irregular times the electrode
responses using a SLS concentration of 0.1 pg/mL. The sensor was stored at 4 °C in a
refrigerator in PBS (pH = 7.2) solution during the stability test. Consequently, the initial
current response of the electrode decreased to 98.48 % after 15 days, whereas 97.16 % of the
original response was retained after 45 days. After 85 days, a response of 92.56 % has been
recorded. Thus, the sensor can be stored for more than 3 months at 4 °C without remarkable
deterioration in its analytical performances. It can be seen that the sensor exhibits remarkable

PT
advantages, such as higher sensitivity, wider linear range and lower detection limit with
satisfactory accuracy suggesting its utilisation for SLS determination in real samples.

RI
3.8. Real samples analysis

SC
The newly developed MIP sensor was tested for the determination of SLS in different
environmental waters and cosmetic products samples, in order to demonstrate its applicability
NU
in practical applications.
For wastewaters and river waters analysis, the results are shown in Table 2. Good RSD
MA

values, ranging between 3.07 % and 7.85 %, were calculated, with good reproducibility,
validating the practicability of the developed sensor for real samples analysis. Therefore, the
environmental waters SLS contents found by the MIP sensor (DPV, EIS) measurements are
D

compared and found to be in good agreement with those of the spectrophotometer used as
E

reference method. Interestingly, it could be stated that this sensor was reliable and effective
PT

for the determination of SLS in environmental waters samples.

In addition, this sensor was applied for the determination of SLS in cosmetic products
CE

samples. The obtained results are summarized in Table 3. The obtained RSD values ranged in
good percent between 0.03 % and 7.38 %. For reliable determination of SLS in the cosmetic
AC

products, the obtained results from the MIP sensor are also compared with those of the
spectrophotometric method. As shown in Table 3, the recorded results from the both methods
are in good accordance. Taking clue to these findings, the whole obtained results essentially
demonstrate that the proposed strategy for SLS measurements based on MIP electrochemical
sensor is accurate, sensitive, consistent and reliable.
A well-known statistical method named partial least square (PLS) has been used in the aim of
extending this study to expect a correlation between the developed MIP sensor and the
spectrophotometer. The underlying objective was to assess if the MIP sensor could predict
SLS contents in environmental waters and cosmetic product samples. The X-block consisted
 ACCEPTED MANUSCRIPT

of the response matrix acquired with the UV-Vis spectrophotometer at 654 nm and Y-block
was a column matrix containing the results of the MIP sensor analysis using DPV technique.
One latent variable and a leave one out cross-validation technique have been used to build the
appropriate models. Fig. 13A and Fig. 13B show the PLS-predicted SLS contents versus the
real SLS values for the environmental waters and cosmetic products samples, respectively.
Calibration plots with good linearity were obtained with satisfactory regression coefficients of
0.99. Due to the good correlation existing between the MIP sensor technique and the
spectrophotometer, the developed MIP sensor could be used as a rapid and alternative way for

PT
SLS content determination in environmental waters and cosmetic products samples.

RI
4. Conclusion

SC
In conclusion, for the first time, a MIP electrochemical sensor for the detection of SLS has
been well developed. It was elaborated via electropolymerization of 2-ATP onto Au-SPE in
NU
the presence of SLS as template molecule. A relatively good linearity was obtained for the
binding of SLS in the adopted concentration range from 0.1 to 1000 pg/mL with high
MA

sensitivity (LOD of 0.18 pg/mL and LOQ of 0.61 pg/mL). The reproducibility, selectivity and
stability of the MIP sensor were all found to be satisfactory. By the means of PLS, the present
MIP sensor demonstrates its capability to be an effective method for the electrochemical
D

determination of SLS in wastewaters, river waters and cosmetic products samples for
E

environmental protection in comparison with the spectrophotometer. Relative to other

PT

techniques, this method has very interesting analytical parameters such as sensitivity with low
detection limit and highly selective separation. On this basis, this strategy can be an
CE

interesting tool and further expected to be used for elaboration of various molecular
imprinting-based sensors for advanced environmental applications.
AC

Acknowledgements
The authors would like to acknowledge Moulay Ismail University for the financial project
‘’Scientific Research Promotion‘’. The authors would like to thank CNRST-Morocco for the
award of the merit scholarship for the PhD student and TROPSENSE under H2020-MSCA-
RISE-2014 project, grant agreement number: 645758 for their support.
 ACCEPTED MANUSCRIPT

References
[1] S. Gupta, A. Pal, P.K. Ghosh, M. Bandyopadhyaya, Performance of waste activated
carbon as a low-cost adsorbent for the removal of anionic surfactant from aquatic
environment, Journal of Environmental Science and Health, Part A 38 (2003) 381-397.

[2] M. Sharma, P. Chadha, Widely used non-ionic surfactant 4-nonylphenol: showing

genotoxic effects in various tissues of Channa punctatus, Environmental Science and
Pollution Research 24 (2017) 11331-11339.

PT
[3] M. Filipović, M. Lukić, V. Krstonošić, S. Đorđević, I. Pantelić, A. Gledović, S. Savić,
Feasibility of a natural surfactant as a stabilizer for cosmetics with liposome-encapsulated

RI
plant stem cells: pre-formulation and formulation through stability studies, Tenside

SC
Surfactants Detergents 53 (2016) 214-226.

[4] D. Myers, Surfactant Science and Technology, 3rd.:ed John Wiley and Sons, New Jersey,
NU
2005.

[5] M. Lechuga, M. Fernández-Serrano, E. Jurado, J. Núñez-Olea, F. Ríos, Acute toxicity of

MA

anionic and non-ionic surfactants to aquatic organisms, Ecotoxicology and environmental

safety 125 (2016) 1-8.
D

[6] Y. Zhao, X.Y. Liu, Y.H. Yang, Sodium dodecyl sulfate and sodium dodecyl sulfonate,
Chem. Life 24 (2004) 277-278.
E
PT

[7] A.B. Caracciolo, M. Cardoni, T. Pescatore, L. Patrolecco, Characteristics and

environmental fate of the anionic surfactant sodium lauryl ether sulphate (SLES) used as
CE

the main component in foaming agents for mechanized tunneling, Environmental

Pollution 226 (2017) 94-103.
AC

[8] C.A. Bondi, J.L. Marks, L.B. Wroblewski, H.S. Raatikainen, S.R. Lenox, K.E. Gebhardt,
Human and environmental toxicity of sodium lauryl sulfate (SLS): evidence for safe use
in household cleaning products, Environmental Health Insights 9 (2015) 27-32.

[9] A. Casari, F. Farnetani, B. De Pace, A. Losi, J.C. Pittet, G. Pellacani, C. Longo, In vivo
assessment of cytological changes by means of reflectance confocal microscopy–
demonstration of the effect of topical vitamin E on skin irritation caused by sodium lauryl
sulfate, Contact Dermatitis 76 (2017) 131-137.
 ACCEPTED MANUSCRIPT

[10] K. Horita, D. Horita, H. Tomita, M. Yasoshima, A. Yagami, K. Matsunaga, Effects of

different base agents on prediction of skin irritation by sodium lauryl sulfate using patch
testing and repeated application test, Toxicology 382 (2017) 10-15.

[11] A. Wibbertmann, I. Mangelsdorf, K. Gamon, R. Sedlak, Toxicological properties and

risk assessment of the anionic surfactants category: Alkyl sulfates, primary alkane
sulfonates, and a-olefin sulfonates, Ecotoxicology and Environmental Safety 74 (2011)
1089-1106.

PT
[12] H. Löffler, R. Happle, Profile of irritant patch testing with detergents: sodium lauryl
sulfate, sodium laureth sulfate and alkyl polyglucoside, Contact Dermatitis 48 (2003) 26-

RI
32.

SC
[13] A. Gavina, S. Bouguerra, I. Lopes, C.R. Marques, M.G. Rasteiro, F. Antunes, T. Rocha-
Santos, R. Pereira, Impact of organic nano-vesicles in soil: The case of sodium dodecyl
NU
sulphate/didodecyl dimethylammonium bromide, Science of the Total Environment 547
(2016) 413-421.
MA

[14] O.R.T. Thomas, G.F. White, Metabolic pathway for biodegradation of sodium dodecyl
sulfate by Pseudomonas sp. C12B, Biotechnology and Applied Biochemistry 11 (1989)
318-327.
D

[15] D.J. Anderson, M.J. Day, N.J. Russell, G.F. White, Die-away kinetic analysis of the
E

capacity of epilithic and planktonic bacteria from clean and polluted river water to
PT

biodegrade sodium dodecyl sulphate, Applied and Environmental Microbiology 56

(1990) 758-763.
CE

[16] T. Ivanković, J. Hrenović, Surfactants in the environment, Arhiv za higijenu rada i

AC

toksikologiju, 61 (2010) 95-109.

[17] D. Yuan, D.Y. Fu, X.S. Zhang, Means and recent development for determination of
anionic surfactants, J. Sichuan Univ. Sci. Eng. (Natur. Sci. Ed.) 18 (2005) 27-32.

[18] E. Ródenas-Torralba, B.F. Reis, Á. Morales-Rubio, M. de la Guardia, An

environmentally friendly multicommutated alternative to the reference method for anionic
surfactant determination in water, Talanta 66 (2005) 591-599.
 ACCEPTED MANUSCRIPT

[19] B. Kovács, B. Csóka, G. Nagy, A. Ivaska, All-solid-state surfactant sensing electrode

using conductive polymer as internal electric contact, Analytica Chimica Acta 437 (2001)
67-76.

[20] M.J. Seguí, J. Lizondo-Sabater, R. Martínez-Máñez, T. Pardo, F. Sancenón, J. Soto, Ion-

selective electrodes for anionic surfactants using a new aza-oxa-cycloalkane as active
ionophore, Analytica Chimica Acta 525 (2004) 83-90.

[21] X. Chen, S. Kang, M.J. Kim, J. Kim, Y.S. Kim, H. Kim, B. Chi, S.J. Kim, J.Y. Lee, J.

PT
Yoon, Thin-film formation of imidazolium-based conjugated polydiacetylenes and their
application for sensing anionic surfactants, Angewandte Chemie International Edition 49

RI
(2010) 1422-1425.

SC
[22] W. Zhang, S.K. Yang, Spectrophotometric determination of anionic surfactant in
wastewater, Journal of Earth Science and Environmental 26 (2004) 92-94.
NU
[23] F. Arduini, L. Micheli, D. Moscone, G. Palleschi, S. Piermarini, F. Ricci, G. Volpe,
Electrochemical biosensors based on nanomodified screen-printed electrodes: Recent
MA

applications in clinical analysis, TrAC Trends in Analytical Chemistry 79 (2016) 114-

126.
D

[24] M. Azimzadeh, M. Rahaie, N. Nasirizadeh, K. Ashtari, H. Naderi-Manesh, An

electrochemical nanobiosensor for plasma miRNA-155, based on graphene oxide and
E

gold nanorod, for early detection of breast cancer, Biosensors and Bioelectronics 77
PT

(2016) 99-106.
CE

[25] K.K. Aswini, A.V. Mohan, V.M. Biju, Molecularly imprinted polymer based
electrochemical detection of L-cysteine at carbon paste electrode, Materials Science and
AC

Engineering: C 37 (2014) 321-326.

[26] A.E. Radi, A.E. El-Naggar, H.M. Nassef, Molecularly imprinted polymer based
electrochemical sensor for the determination of the anthelmintic drug oxfendazole,
Journal of Electroanalytical Chemistry 729 (2014) 135-141.

[27] M.H. Do, A. Florea, C. Farre, A. Bonhomme, F. Bessueille, F. Vocanson, N. Jaffrezic-

Renault, Molecularly imprinted polymer-based electrochemical sensor for the sensitive
detection of glyphosate herbicide, International Journal of Environmental Analytical
Chemistry 95 (2015) 1489-1501.
 ACCEPTED MANUSCRIPT

[28] X. Kan, Z. Xing, A. Zhu, Z. Zhao, G. Xu, C. Li, H. Zhou, Molecularly imprinted
polymers based electrochemical sensor for bovine hemoglobin recognition, Sensors and
Actuators B: Chemical 168 (2012) 395-401.

[29] K. Haupt, Molecularly imprinted polymers in analytical chemistry, Analyst 126 (2001)
747-756.

[30] L. Li, L. Yang, Z. Xing, X. Lu, X. Kan, Surface molecularly imprinted polymers-based
electrochemical sensor for bovine hemoglobin recognition, Analyst 138 (2013) 6962-

PT
6968.

RI
[31] P. Liu, X.H. Zhang, W. Xu, C.H. Guo, S. Wang, Electrochemical sensor for the
determination of brucine in human serum based on molecularly imprinted poly-o-

SC
phenylenediamine/SWNTs composite film, Sensors Actuators B : Chemical 163 (2012)
84-89.
NU
[32] X. Li, Y. Wang, X. Yang, J. Chen, H. Fu, T.C. Yo, Conducting polymers in
environmental analysis, TrAC Trends in Analytical Chemistry 39 (2012) 163-179.
MA

[33] C. Malitesta, E. Mazzotta, R. A. Picca, A. Poma, I. Chianella, S.A. Piletsky, MIP

sensors–the electrochemical approach, Analytical and Bioanalytical Chemistry 402
(2012) 1827-1846.
D

[34] P.S. Sharma, A.P. Le, F. D’souza, W. Kutner, Electrochemically synthesized polymers in
E

molecular imprinting for chemical sensing, Analytical and Bioanalytical Chemistry 402
PT

(2012) 3177-3204.
CE

[35] M. Saber-Tehrani, A. Pourhabib, S.W. Husain, M. Arvand, A simple and efficient

electrochemical sensor for nitrite determination in food samples based on Pt nanoparticles
AC

distributed poly(2-aminothiophenol) modified electrode, Food Analytical Methods 6

(2013) 1300-1307.

[36] C. Xue, Q. Han, Y. Wang, J. Wu, T. Wen, R. Wang, J. Hong, X. Zhou, H. Jiang,
Amperometric detection of dopamine in human serum by electrochemical sensor based on
gold nanoparticles doped molecularly imprinted polymers, Biosensors and Bioelectronics
49 (2013) 199-203,

[37] X. Sun, L. Zhang, H. Zhang, H. Qian, Y. Zhang, L. Tang, Z. Li, Development and
application of 3-chloro-1,2-propandiol electrochemical sensor based on a
 ACCEPTED MANUSCRIPT

polyaminothiophenol modified molecularly imprinted film, Journal of Agricultural and

Food Chemistry 62 (2014) 4552-4557.

[38] B.B. Prasad, I. Pandey, A. Srivastava, D. Kumar, M.P. Tiwari, Multiwalled carbon
nanotubes-based pencil graphite electrode modified with an electrosynthesized
molecularly imprinted nanofilm for electrochemical sensing of methionine enantiomers,
Sensors and Actuators B: Chemical 176 (2013) 863-874.

[39] F.T.C. Moreira, S. Sharma, R.A. Dutra, J.P. Noronha, A.E. Cass, M.G.F. Sales, Protein-

PT
responsive polymers for point-of-care detection of cardiac biomarker, Sensors and
Actuators B: Chemical 196 (2014) 123-132.

RI
[40] D.R. Albano, F. Sevilla III, Piezoelectric quartz crystal sensor for surfactant based on

SC
molecularly imprinted polypyrrole, Sensors and Actuators B: Chemical, 121 (2007) 129-
134. NU
[41] X. Hao, J.L. Lei, N. Bing Li, H.Q. Luo, An electrochemical sensor for sodium dodecyl
sulfate detection based on anion exchange using eosin Y/polyethyleneimine modified
MA

electrode, Analytica Chimica Acta 852 (2014) 63-68.

[42] M. Riskin, R. Tel-Vered, T. Bourenko, E. Granot, I. Willner, Imprinting of molecular

recognition sites through electropolymerization of functionalized Au nanoparticles:
D

development of an electrochemical TNT sensor based on π-donor− acceptor interactions,

E

Journal of the American Chemical Society 130 (2008) 9726-9733.

PT

[43] L.M. Popescu, A. Meghea, R.M. Piticescu, E. Vasile, High pressure synthesis procedure
to obtain nanostructured composites for regenerative medicine, Journal of Optoelectronics
CE

and Advanced Materials 9 (2007) 3354-3357.

[44] M. Koga, Y. Yamamichi, Y. Nomoto, M. IRIE, T. TANIMURA, T. YOSHINAGA,

AC

Rapid determination of anionic surfactants by improved spectrophotometric method using

methylene blue, Analytical Sciences 15 (1999) 563-568.

[45] S. Devi, M.C. Chattopadhyaya, Determination of sodium dodecyl sulfate in toothpastes

by a PVC matrix membrane sensor, Journal of Surfactants and Detergents 16 (2013) 391-
396.

[46] Ü. Şengül, Comparing determination methods of detection and quantification limits for
aflatoxin analysis in hazelnut, Journal of Food and Drug Analysis 24 (2016) 56-62.
 ACCEPTED MANUSCRIPT

[47] A.N. Reshetilov, I.N. Semenchuk, P.V. Iliasov, L.A. Taranova, The amperometric
biosensor for detection of sodium dodecyl sulfate, Analytica Chimica Acta 347 (1997)
19-26.
[48] A.R. Ghiasvand, Z. Taherimaslak, M. Allahyari, Sensitive and selective
spectrophotometric and a new adsorptive stripping voltammetric determination of sodium
dodecyl sulfate in nonaqueous solution after its extraction using toluidine blue,
International Journal of Electrochemical Science 4 (2009) 320-335.
[49] Q.Z. Zhai, Catalytic kinetic spectrophotometric determination of sodium dodecyl

PT
sulphonate, Instrumentation Science and Technology 38 (2010) 135-142.

RI
SC
NU
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Figure captions
Fig. 1. Schematic of the portable device system and the spectrophotometer used for
electrochemical measurements.
Fig. 2. Schematic illustration of the elaboration of the MIP sensor.
Fig. 3. Optimization of experimental parameters for MIP sensor: (A) 2-ATP deposition time,
(B) Number of cycle for the electropolymerization, (C) Extraction time and (D) Incubation
time.
Fig. 4. CVs for the electropolymerization MIP in [Fe (CN)6]3-/4-. Potential range from -0.35 to

PT
0.8 V; Scan rate: 100 mV/s; number of scans: 10.
Fig. 5. (A) Cyclic voltammograms and (B) Nyquist plots of 5 mM [Fe (CN)6]3-/4- solution at

RI
Au-SPE/MIP and Au-SPE/NIP.

SC
Fig. 6. (A) AFM image side view, (B) AFM-2D topographic image and (C) AFM-3D images
of electropolymerization step.
NU
Fig. 7. (A) AFM image side view, (B) AFM-2D topographic image and (C) AFM-3D images
of extraction step.
Fig. 8. FTIR spectra of: (A) Electropolymerization step and (B) SLS extraction step.
MA

Fig. 9. (A) Differential pulse voltammograms and (B) Nyquist plots at different SLS
concentrations towards MIP sensor.
D

Fig. 10. (A) Differential pulse voltammograms and (B) Nyquist plots at different SLS
concentrations towards NIP sensor.
E

Fig. 11. Calibration curves obtained for MIP and NIP sensor by: (A) DPV and (B) EIS.
PT

Fig. 12. MIP sensor responses for interfering molecules.

Fig. 13. Results of PLS prediction of SLS levels in (A) Environmental waters and (B)
CE

Cosmetic products by means of DPV method versus spectrophotometric method.

AC
 ACCEPTED MANUSCRIPT

Table 1. Comparison of the developed sensor for SLS detection with other methods

Method Linear range Detection limit Reference

(pg/mL) (pg/mL)
Anion-exchange-based
1×106 – 4×106 9×105 [41]
amperometry

Amperometric biosensor 7.5×105 – 109 2.5×105 – 7.5×105 [47]

PT
Spectrophotometric 5×104 – 4×106 3.3×104
[48]
And voltammetric methods 5×104 – 1.75×107 1.2×103

RI
Catalytic Kinetic
Spectrophotometric 3×105 – 1×106 1.52×105 [49]

SC
Determination
Molecularly imprinted sensor Present
0.1 – 1000 0.18
Au-SPE/MIP work
NU
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 2. Determination of SLS in environmental waters samples

MIP sensor Spectrophotometer
Environmental RSD RSD RSD
DPV EIS Spectro
waters samples (%) (%) (%)
(105 ng/mL) (n=3) (105 ng/mL) (n=3) (105 ng/mL) (n=3)

Wastewater 1 /
63.62 7.67 63.26 5.87 63.54 9.35
Khemisset
River water 1 /
14.29 7.26 14.15 7.85 14.25 7.33

PT
Fes
River water 2/
0.18 3.45 0.18 3.65 0.18 4.86

RI
Fes
Wastewater 2/

SC
211.63 3.85 209.30 2.26 211.33 3.68
Rabat
River water 3/
5.13 3.38 4.93 4.58 5.03 2.58
NU
Fes
River water 4/
16.42 3.07 16.69 3.14 16.76 3.16
Sidi kacem
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 3. Determination of SLS in cosmetic products samples

MIP sensor Spectrophotometer
Cosmetic
products DPV RSD EIS RSD Spectro RSD
3 3
samples (10 ng/mL) (%) (10 ng/mL) (%) (103 ng/mL) (%)
(n=3) (n=3) (n=3)

Shampoo 1 5.25 0.10 5.21 0.09 5.04 0.30

Shampoo 2 0.86 0.95 0.86 0.55 0.83 0.60

PT
Toothpaste 77.55 0.03 77.53 0.09 73.26 3.85

RI
Shower gel 2.65 5.38 2.31 7.38 2.54 8.20

SC
NU
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Highlights

 An electrochemical molecularly imprinted sensor for SLS detection is proposed.

 The sensor shows high sensitivity and good selectivity.
 The imprinted sensor exhibits good reproducibility and long-term storage.
 A lower LOD of 0.18 pg/mL is achieved in this study compared to reported works.
 Obtainment of satisfactory sensor results for environmental waters and cosmetic

PT
products samples analysis correlated with the spectrophotometric method’s.

RI
SC
NU
MA
E D
PT
CE
AC
Graphics Abstract
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13