Mawin J Martinez Jimenez1,2, Alba Avila3, Anerise de Barros4, Elvis Oswaldo Lopez5, Fernando Alvarez2, Antonio Riul2, Jaime Andres Perez-Taborda1,3. 1. Colombian Society of Engineering Physics (SCIF), Pereira 660003, Colombia. 2. Department of Applied Physics, "Gleb Wataghin" Institute of Physics, University of Campinas-UNICAMP, Campinas 13083-970, São Paulo, Brazil. 3. Centro de Microelectrónica (CMUA), Departamento de Ingeniería Eléctrica y Electrónica, Universidad de los Andes, Bogotá 111711, Colombia. 4. Laboratory of Functional Materials, Institute of Chemistry, University of Campinas-UNICAMP, P.O. Box 6154, Campinas 13083-970, São Paulo, Brazil. 5. Department of Experimental Low Energy Physics, Brazilian Center for Research in Physics (CBPF), Rua Dr. Xavier Sigaud 150, Rio de Janeiro 22290-180, Brazil.
Abstract
Heavy metal pollution is posing a severe health risk on living organisms. Therefore, significant research efforts are focused on their detection. Here, we developed a sensing platform sensor for the selective detection of lead(II) acetate. The sensor is based on self-assembled polyethyleneimine-functionalized carbon nanotubes (PEI-CNTs) and graphene oxide films deposited onto gold interdigitated electrodes. The graphene-based nanostructure showed a resistive behavior, and the fabricated layer-by-layer film was used to detect Pb(II) acetate in an aqueous solution by comparison of three electrochemical methods: impedance spectroscopy, amperometry, and potentiometry stripping analysis. The results obtained from different methods show that the detection limit was down to 36 pmol/L and the sensitivity up to 4.3 μAL/μmol, with excellent repeatability. The detection mechanism was associated with the high affinity of heavy metal ions with the functional groups present in the PEI-CNTs and GO, allowing high performance and sensitivity. The achieved results are important for the research toward integrated monitoring and sensing platforms for Pb(II) contamination in drinking water.
Heavy metal pollution is posing a severe health risk on living organisms. Therefore, significant research efforts are focused on their detection. Here, we developed a sensing platform sensor for the selective detection of lead(II) acetate. The sensor is based on self-assembled polyethyleneimine-functionalized carbon nanotubes (PEI-CNTs) and graphene oxide films deposited onto gold interdigitated electrodes. The graphene-based nanostructure showed a resistive behavior, and the fabricated layer-by-layer film was used to detect Pb(II) acetate in an aqueous solution by comparison of three electrochemical methods: impedance spectroscopy, amperometry, and potentiometry stripping analysis. The results obtained from different methods show that the detection limit was down to 36 pmol/L and the sensitivity up to 4.3 μAL/μmol, with excellent repeatability. The detection mechanism wasassociated with the high affinity of heavy metal ions with the functional groups present in the PEI-CNTs and GO, allowing high performance and sensitivity. The achieved results are important for the research toward integrated monitoring and sensing platforms for Pb(II)contamination in drinking water.
Non-essential
heavy metals such asmercury (Hg), arsenic (As),
lead (Pb), and cadmium (Cd) are cumulative, non-biodegradable, and
highly toxic, causing serious health problems on living organisms
as they are prevalent even at very low concentrations.[1] Briefly, Pb and Pb-based compounds are widely known as
hazardous environmental pollutants and toxic agents that can cause
severe environmental damage and diseases in animals and humans.[2−5] Moreover, Pb ions are a common soil contaminant that causes a decrease
in soil biological activity.[6] Pb ions can
enter the human body through inhalation, ingestion, or consumption
of contaminated food and water and accumulate in different tissues
and distinct vital organs. Several diseases, such as those associated
with memory deterioration in children, reduced fertility in men, intellectual
deficit, adverse effects on the central nervous, gastrointestinal,
and immune systems, and producing mitochondrial degradation and DNA
damage, have been reported.[7−9] Lead acetate (PbAc), also known
aslead (II) acetate, is a lead-based component widely used in the
processing of various commercial products. For example, PbAc is used
to manufacture cosmetics, pesticides, water repellents, mildew protection
products, and so on.[10,11] On the whole, several adverse
health problems in living beings are associated with prolonged exposure
to lead acetate.[12,13] Consequently, researchers are
highly keen on developing new materials that are highly sensitive
and capable of determining these ions in an aqueous solution.In recent years, graphene-based nanostructures appeared as an emerging
class of materials in many applications, from environmental monitoring
to human health issues.[14,15] Graphene is an atomic
monolayer of sp2 hybridized carbon atoms arranged in a
hexagonal pattern and considered a semi-metallic material with zero
band gap, high electronic mobility (200,000 cm2 V–1 s–1), surface-to-volume ratio (2600 m2 g–1), optical transmission (∼97.7%), mechanical
strength (130 GPa), and high electrical and thermal conductivities.[16,17] It naturally arouses interest in different areas; however, pristine
graphene is not favorable for sensing applications despite the exceptional
properties.[18] In this sense, various approaches
have been employed to functionalize graphene-based materials to expand
sensing platforms’ applications.Graphene-based materials
modified with organiccompounds containing
functional groups have the natural ability to improve the mechanisms
of adsorption of heavy metal ions through electrostatic forces, chelation,
and hydrogen bonds, thus facilitating their detection. Graphene oxide
(GO) and carbon nanotubes (CNTs) are of graphene-based materials,
with GO having differently oxygenated groups (hydroxyl, epoxide, carbonyl,
and carboxyl) at the edges and in the basal plane, making it an advantageous
material to form stable solutions in water and other organic solvents.[19] These oxygenated groups serve as active sites
for adsorbing heavy metal ions, making GO a strong candidate for sensing
due to adsorption of various heavy metal pollutants.[20] In contrast, CNTs have extraordinary hydrophobic performance,
it being impossible to study their properties through solution-based
techniques.[21] Recently, polyethyleneimine
(PEI), a polymericcation rich in amine groups that contains a nitrogen
atom with a lone pair, demonstrated outstanding ability in the functionalization
of CNTs. Besides, PEI is widely known for its chelating metal properties;
after the functionalization of CNTs with PEI, it allows the possibility
of introducing active sites in CNTs for heavy metal adsorption.[22] Polyethyleneimine-functionalized carbon nanotubes
(PEI-CNTs) have shown efficiency in technological applications such
as solar cells,[23] removal of pollutants,[24] and an excellent electrochemical sensing platform
for silver nanoparticles,[25] chromium ions,[26] and promethazine hydrochloride.[27] Nevertheless, it has been little explored for detecting
heavy metal ions and even less regarding the PEI-CNT/GOcomposites.
In particular, Huang et al. reported an electrochemical sensor using
a glassy carbon electrode (GCE) modified with three-dimensional GO-CNT
hybrid nanocomposite for the simultaneous detection of Pb2+ and Cd2+ ions using anodic stripping voltammetry.[28] Shuping Wu et al. developed an electrochemical
sensing platform integrating a hierarchical nanostructure (CNT/graphene/gold
nanoparticle (AuNP)/Nafion) on the surface of a GCE electrode for
the sensitive detection of Pb2+ in water and milk samples,
exhibiting rapid response and high sensitive Pb2+ detection.[29] In contrast, different nanocomposites containing
graphene-based materials have been widely used to determine trace
amounts of heavy metal ions.[30,31] Hui et al. have developed
an electrochemical sensor for the detection of copper (Cu) and zinc
(Zn) ions based on a Au electrode modified through layer-by-layer
(LbL) assembly of titanium carbide (Ti3C2T) and CNT composites. The sensitivity of
the sensor was attributed to a synergistic effect between Ti3C2T and CNTs, which decreases
the resistance of electron exchange during the redox process.[32] Zhou et al. fabricated a field-effect transistor
(FET) device-based sensor to specifically detect Pb2+ ions
in water using a reduced graphene oxide (rGO)/AuNP hybrid structure,
rGO being used as the semiconducting channel material and AuNPs being
used to improve the performance of the sensor.[33]In this study, we report a highly sensitive platform
for detecting
Pb(II) acetate in an aqueous solution. An easy strategy was implemented
to prepare a PEI-CNT, designed to be a positive polyelectrolyte. Our
functionalization process is based on wrapping CNTs with a PEI polymericchain. It has been shown that PEIcan act as a good supramolecular
complex to adsorb on the CNT’s surface via intermolecular forces.
Therefore, it paves the way to manufacture a nacre-like structure
with GO onto gold interdigitated electrodes (IDEs) in a straightforward,
versatile, low-cost, and functional way through the LbL technique.[34] PEI-CNTs were characterized by UV–vis
and XPS spectroscopy. The fabricated sensor was used to detect Pb(II)acetate in an aqueous solution by impedance spectroscopy (IS), amperometry,
and potentiometric stripping analysis (PSA). The detection mechanism
wasassociated with the high affinity of heavy metal ions with functional
groups present in the PEI-CNTs and GO, allowing high performance and
sensitivity.
Results and Discussion
Stable Aqueous Solution of the PEI-CNT Composite
The
CNT functionalization process is based on a supramolecular
complexation method using PEI to wrap the CNTs, as illustrated in Figure a. A black water-stable
solution is obtained, remaining stable after 15 days of preparation
(see Figure b). It
is a non-covalent approach to functionalize CNTs through π–π
interactions, hydrogen bonding, and van der Waals forces.[35] In this way, it is possible to synergistically
combine PEI-CNTs with GO for high affinity with heavy metal ions,[36] thus allowing an increase in the sensitivity
of sensors through the strong electrostatic interaction with Pb ions
in lead acetate.
Figure 1
Aqueous solution of the PEI-CNT composite: (a) Schematic
representation
of the cationic PEI-functionalized CNT. (b) Picture of the as-prepared
PEI-CNT solution (left) and picture taken after preparation for 15
days for the solutions of PEI-functionalized CNTs (right).
Aqueous solution of the PEI-CNT composite: (a) Schematic
representation
of the cationicPEI-functionalized CNT. (b) Picture of the as-prepared
PEI-CNT solution (left) and picture taken after preparation for 15
days for the solutions of PEI-functionalized CNTs (right).
UV–Vis and XPS Characterizations of
the PEI-CNT Composite
The successful functionalization of
CNTs with PEI is certified by UV–vis and XPS spectroscopy. Figure a shows the UV–vis
absorption spectra obtained from the aqueous dispersions of GO, PEI-CNTs,
and PEI. The spectrum of the GO aqueous solution displays a characteristic
absorption band at 230 nm characteristic of the π–π*
transitions due to the C–C bonds from the sp2 hybridization
regions, and a small shoulder at 300 nm due to the n−π*
transitions of C=O bonds present in sp3 hybridizations.[37] The UV–vis spectrum of the as-prepared
PEI-CNT suspensions shows an absorbance band around 270 nm, commonly
attributed to the π–π* transitions of the aromaticC–C bonds of a graphitic structure.[38] Besides, it can be seen from Figure a that the complexation of CNTs does not inhibit the
PEI absorption response.
Figure 2
(a) UV–vis absorption spectra of GO aqueous
solution, PEI-functionalized
CNTs, and PEI. (b) XPS survey spectra of unmodified CNTs and PEI-functionalized
CNTs. High-resolution C 1s core-level spectra of (c) unmodified CNTs
and (d) PEI-functionalized CNTs. (e) High-resolution N 1s spectrum
of PEI-functionalized CNTs.
(a) UV–vis absorption spectra of GO aqueous
solution, PEI-functionalized
CNTs, and PEI. (b) XPS survey spectra of unmodified CNTs and PEI-functionalized
CNTs. High-resolution C 1s core-level spectra of (c) unmodified CNTs
and (d) PEI-functionalized CNTs. (e) High-resolution N 1s spectrum
of PEI-functionalized CNTs.The XPS data were recorded before and after the functionalization
of CNTs with PEI. Figure b shows the XPS survey spectra of CNTs and PEI-CNTs, showing
relatively intense peaks that indicate the presence of C and O in
both non-functionalized and functionalized samples. At the same time,
N is only present in PEI-CNTs. Additional peaks are observed in the
survey XPS spectrum of PEI-CNTs generated from the Si substrate. The
high-resolution band associated with the electron C 1s core-level
of unmodified CNTs and PEI-functionalized CNTs are shown in Figure c,d, respectively.
The band related to the C 1s electron of the unmodified CNTs displays
the main peak near 284.3 eV, which can be resolved mainly by deconvolution
into two peaks, generally attributed to sp2 (284.2 eV)
and sp3 (284.9 eV) hybridized graphiticcarbon atoms.[39] The peak at 286 eV can be due to C–O–C
bonds from impurities. The C–C maximum peak of PEI-CNTs shows
a significant up-shift from 284.3 to 285.0 eV compared to that of
CNTs (Figure d), indicating
rehybridization of sp2 carbon to sp3 carbon.[40] It was deconvoluted into four peaks at 284.3,
285.0, 285.7, and 286.9 eV, commonly assigned to the sp2, sp3, C–N amine, and −N–C=O
bondings, respectively.[41] After PEI functionalization,
an additional peak appears in the N 1s region, showing PEIs presence
after in the surface modification of the wrapped CNTs. Deconvolution
of the N 1s signal reveals peaks at 399.9, 400.8, and 402.1 eV, associated
with interactions between the CNT surface and amine groups through
the N–C, N=C, and N–C=O bonds, respectively.
LbL Self-Assembled Multilayer Films of PEI-CNTs
and GO and Their Electrical Properties
The LbL assembly was
employed in the growth of the film onto IDEs, consisting of spontaneous
organization and sequential adsorption of PEI-CNTs and GO. It is mainly
driven by van der Waals forces, hydrogen bonds, π–π
stacking, and electrostatic interactions.[42]Figure a illustrates
the scanning electron microscopy (SEM) images of as-deposited (PEI-CNTs/GO)25 films onto Au-IDEs. As shown in Figure b, the coating film is all over the IDE region,
composed of randomly clustered GO nanoplatelets (see Figure c,d). The electrical properties
of (PEI-CNTs/GO)25 multilayered films were analyzed by I–V characteristiccurves and electrical
impedance measurements (see Figure ). In Figure a, it can be seen that the electrical I–V response of the (PEI-CNTs/GO)25 LbL film has
linearity similar to that of a resistor behavior. In this case, electrical
impedance is a powerful tool to better understand the LbL structure’s
behavior. The real and imaginary parts of impedance are associated
with an Ohmic resistance (Z′ = R) and capacitance (Z″ = 1/jωC), respectively, and related through Z = Z′ + Z″.
It is well-known that the phase (θ) represents the angle between
impedance components in the complex plane (inset Figure b), giving −90, 0, or
90° when the electrical response presents a capacitive, resistive,
or inductive behavior, respectively.[43]Figure b shows the Bode
plots for the phase before and after the film nanostructuration on
the IDEs. Bare IDEs have a phase of −90° in almost the
whole frequency regions analyzed, and after the LbL formation, it
becomes approximately 0° from 1 to 10 kHz. For frequencies above
10 kHz, the response is not purely resistive because of the capacitive
effects from the design of the IDEs. Figure c reinforces the idea of a strictly resistive
behavior to (PEI-CNTs/GO)25 LbL films in the same frequency
region of θ = 0. Similarly, the capacitive behavior of the bare
IDEs is shown in Figure d in the frequency range where θ = −90.
Figure 3
SEM images of the as-deposited
(PEI-CNT/GO) film onto IDEs. (a)
Representative top view of the LbL film onto Au-IDEs. (b) Zoomed view
of the image (a). Surface morphology of the LbL film: (c) on the Au
finger and (d) between a finger pair.
Figure 4
Room-temperature
electrical characterization of the (PEI-CNT/GO)[25] LbL film. (a) Current–potential applied
plot, the measurement was performed at 50 mV/s. IS was performed from
1 Hz to 1 MHz at an amplitude of 25 mV. (b–d) Corresponding
Bode phase, Impedance Real, and Capacitance diagrams, respectively.
SEM images of the as-deposited
(PEI-CNT/GO) film onto IDEs. (a)
Representative top view of the LbL film onto Au-IDEs. (b) Zoomed view
of the image (a). Surface morphology of the LbL film: (c) on the Au
finger and (d) between a finger pair.Room-temperature
electrical characterization of the (PEI-CNT/GO)[25] LbL film. (a) Current–potential applied
plot, the measurement was performed at 50 mV/s. IS was performed from
1 Hz to 1 MHz at an amplitude of 25 mV. (b–d) Corresponding
Bode phase, Impedance Real, and Capacitance diagrams, respectively.So far, we have shown a straightforward approach
through a non-covalent
method for CNT functionalization using PEI. Also, the combination
of the individual properties of PEI-CNTs and GO through the LbL technique
is advantageous for the formation of a graphene-based nanostructure
with suitable and ideal properties for a sensing platform.
(PEI-CNT/GO) LbL Film as a Platform Sensing
of Pb(II) Acetate
Different strategies were approached to
study the capacity of the (PEI-CNT/GO)25 LbL film as a
detection platform. IS and amperometry were used to investigate the
LbL film’s ability to operate as an impedimetric or chemiresistive
sensor to detect different analytes and distinguish different lead
acetateconcentrations. Additionally, the PSA was performed aiming
at the specific detection of Pb(II) acetate.The detection mechanism
of an impedimetric sensor is based on changes produced in the sensing
layer and detected through impedance measurements.[44] A close comparison of the data obtained from impedance
measurements is shown in Figure . The impedance magnitude and the real and complex
parts of the impedance show the device’s ability to differentiate
all samples analyzed (see Figure a–c). A clear difference between impedimetric
responses is observed from 1 kHz to 1 MHz. The Nyquist plot shows
a tendency to form a semicircle (Figure d). Where the diameter of the formed semicircle
can be related to the film’s electrical resistance. Therefore,
the decrease in the semicircle can be related to the reduction in
the interfacial electrical resistance. When ioniccharges in the solution
increase, the electrical resistance of the (PEI-CNT/GO)25 LbL film decreases due to the affinity between ions in the solution
and functional groups (i.e., hydroxyl, carboxyl, epoxy, carboxyl,
and amine groups) in both GO and PEI-CNTs. Ions are attracted to the
solid–liquid interface, and the charge transport properties
of the LbL film are then strongly altered.
Figure 5
IS plots of the (PEI-CNT/GO)25 LbL film submerged in
ultrapure water, 0.1 M HCl, and 7.4 μM Pb(II) acetate in 0.1
M HCl solution. (a) Impedance magnitude vs frequency, (b) real part
vs frequency, (c) imaginary part vs frequency, and (d) Nyquist plots.
IS plots of the (PEI-CNT/GO)25 LbL film submerged in
ultrapure water, 0.1 M HCl, and 7.4 μM Pb(II) acetate in 0.1
M HCl solution. (a) Impedance magnitude vs frequency, (b) real part
vs frequency, (c) imaginary part vs frequency, and (d) Nyquist plots.Impedance measurements were also taken by successively
adding a
100 μmol/L Pb(II) acetate stock solution into the 0.1 mol/L
HCl solution under continuous stirring. The impedimetric responses
obtained are displayed in Figure a for Z/Z0, with Z being the impedance module obtained for
a certain Pb(II) acetateconcentration and Z0 the sensor response 0.1 mol/L HCl solution. The lowest measured
concentration was 100 nmol/L, without significant changes observed
in the impedimetric response of the sensor at lower concentrations. Figure b–d displays
calibration plots using the data obtained from the impedance module,
the real and imaginary impedance at fixed 10 kHz. Figure b,c shows an excellent hyperboliccorrelation coefficient (R = 0.99) for the both impedance
module and real impedance. The imaginary impedance exhibits a linear
behavior (R = 0.96) from 1 to 7.4 μmol/L and
hyperbolic below 1 μmol/L (see Figure and inset Figure d). It is important to highlight that the
impedimetric sensor’s sensitivity is derived from the affinity
of functional groups in the sensing layer with ioniccharges in the
analyte. The oxygen- and nitrogen-containing functional groups, such
as hydroxyl (−OH), carboxyl (−COOH), epoxy (−COC−),
carbonyl (−C=O), and amine (N–H), contribute
to improved sorption performance and serve as active sites for Pb
ion adsorption. This strongly affects the electrical properties and
the charge carrier mechanisms in the (PEI-CNT/GO)25 LbL
film. Consequently, the detection of Pb(II) acetate is facilitated
in impedance measurements.
Figure 6
Impedimetric response of the (PEI-CNT/GO)25 LbL film
for the successive addition of Pb(II) acetate in 0.1 mol/L HCl solution.
(a) Relative impedance modulus (Z/Z0) vs frequency. (b–d) Calibration plots obtained
from the impedance module, real and imaginary impedance at a fixed
frequency value of 10 kHz. The inset of (d) shows the behavior of Z″ at concentrations below 1 μmol/L.
Impedimetric response of the (PEI-CNT/GO)25 LbL film
for the successive addition of Pb(II) acetate in 0.1 mol/L HCl solution.
(a) Relative impedance modulus (Z/Z0) vs frequency. (b–d) Calibration plots obtained
from the impedance module, real and imaginary impedance at a fixed
frequency value of 10 kHz. The inset of (d) shows the behavior of Z″ at concentrations below 1 μmol/L.The detection mechanism of chemiresistive sensors
is based on changes
in the electrical conductivity of the sensing layer when the concentration
of impurities in the aqueous solution is modified.[45]Figure shows the amperometriccurrent response of a (PEI-CNT/GO)25 chemiresistor obtained after the successive addition of Pb(II) acetate
stock solution in 0.1 mol/L HCl under continuous stirring. The figures
also compare previous results with the electrical response obtained
in ultrapure water (black line) and 0.1 mol/L HCl (red line). The
voltage was set at −0.3, −0.4, −0.6, and −0.8
V to find the most sensitive condition and ensure an optimal chemiresistive
sensor performance (see Figure a–d). The choice of negative voltages is due to the
stripping peak current of lead ions in that region.[46] For all voltages analyzed, a clear difference was observed
in the electrical response obtained in pure water and 0.1 mol/L HCl
solution. The significant changes between measurements made by varying
the Pb(II) acetateconcentration from 500 nmol/L to 7.4 μmol/L
are more noticeable from −3 V. Moreover, the difference in
the amperometriccurrent response increases as the voltage becomes
more negative. This is attributed to improvements in the kinetics
of Pb(II)cations attracted to the interface at negative voltages
and also to its affinity with oxygenated groups in GO. Figure e shows the amperometriccurrent
response of the chemiresistive sensor as the Pb(II) acetateconcentration
increases, with the current fixed for a particular time after the
measurements stabilize. A linear behavior can be observed between
the electriccurrent and the Pb(II) acetateconcentration at all applied
voltages. When comparing the current–concentration plots corresponding
to each voltage, it can be seen that the difference lies in the slope
of each figure, being greater as the voltage becomes more negative. Figure f illustrates the
sensitivity versus Voltage plot; the straight-line response was obtained
between −0.3 and −0.8 V, obtaining a sensitivity of
up to 4.3 μAL/μmol, comparable to the values reported
in the literature.[47]
Figure 7
Amperometric response
of the (PEI-CNT/GO)25 LbL film
for the successive addition of lead(II) acetate in 0.1 mol/L HCl solution.
The voltages of (a) −0.3, (b) −0.4, (c) −0.6,
and (d) −0.8 V were applied to evaluate the performance of
the device in the detection of Pb(II) acetate. (e) Stabilized current
response vs concentration of Pb(II) acetate for −0.3, −0.4,
−0.6, and −0.8 V. (f) Sensitivity vs applied voltage
of chemiresistive sensor.
Amperometric response
of the (PEI-CNT/GO)25 LbL film
for the successive addition of lead(II) acetate in 0.1 mol/L HCl solution.
The voltages of (a) −0.3, (b) −0.4, (c) −0.6,
and (d) −0.8 V were applied to evaluate the performance of
the device in the detection of Pb(II) acetate. (e) Stabilized current
response vs concentration of Pb(II) acetate for −0.3, −0.4,
−0.6, and −0.8 V. (f) Sensitivity vs applied voltage
of chemiresistive sensor.Moreover, the PSA technique needs concentrations above 500 μmol/L
to get a response from the device, and then we set a deposition potential
at −0.8V during 30s. With the PSA technique, it was necessary
to use concentrations above 500 μmol/L to obtain a response
from the device. Figure shows the PSA measurements for the successive increase in the concentration
of Pb(II) acetate under stirring. A well-defined stripping signal
was found around −0.64 V, associated with the Pb2+ ions present in Pb(II) acetate.[48] The
analytical signal was an exponential function of the Pb(II) acetateconcentration within the 520–1200 μmol/L range, with
a correlation coefficient of R = 0.98 (inset Figure ). These results
suggest strong interactions between Pb(II) acetate and the (PEI-CNT/GO)25 LbL film, which might be the origin for the sensitivity
of the layer sensing to Pb(II) acetate due to the electrostatic attraction
of Pb2+ ions to the solid–liquid interface.
Figure 8
Stripping potentiograms
for increasing levels of Pb(II) acetate
in the 0.1 mol/L HCl solution. Analytical curve (inset) for Pb(II)
acetate. Exponential behavior covering the range of concentration
from 550 to 1250 μmol/L.
Stripping potentiograms
for increasing levels of Pb(II) acetate
in the 0.1 mol/L HCl solution. Analytical curve (inset) for Pb(II)acetate. Exponential behavior covering the range of concentration
from 550 to 1250 μmol/L.
Analytical Performance of the Proposed Sensing
Platform
So far, three different techniques have been used
to detect Pb ions. Then, the performance of the sensor will be assessed
from the limit of detection (LOD). The LOD wascalculated by taking
a signal-to-noise ratio of 3, with calculations presented in the Supporting
Information. The corresponding calibration plots displayed good linearity
between the electrical response and the Pb(II) acetateconcentration
from the different methods used here, with a correlation coefficient
of 0.9999 (see Figure S1). The LODs for
impedimetric and amperometry sensors for Pb ions were 36 pmol/L and
78 nmol/L, respectively. When using the PSA technique, the LOD for
Pb ions was ∼25 μmol/L.Under optimal experimental
conditions, the (PEI-CNT/GO)25 LbL film was highly sensitive
(LOD = 36 pmol/L) toward Pb ions and significantly lower than the
guideline value of 72 nmol/L for drinking water given by the Environmental
Protection Agency.[49] A comparison of the
performance of Pb ion detection using different techniques and materials
is summarized in Table . It can be seen that the proposed sensing platform has advantages
in the detection of Pb ions.
Table 1
Comparison of the
Analytical Performance
of Different Methods and Materials for the Detection of Pb Ions
detection
method
materials
LOD
refs
differential pulse voltammetry
MOFs HKUST-1
2 nmol/L
(50)
cyclic
voltammetry
MOF-177
4 nmol/L
(51)
anodic
stripping voltammetry
copper
21 nmol/L
(52)
differential
pulse voltammetry
polypyrrole NPs
55 nmol/L
(53)
differential
pulse stripping
voltammetry
graphene aerogel/MOF composites
1.5 nmol/L
(54)
square wave anodic stripping
voltammetry
rGO/graphite-C3N4 composite
1.07 pmol/L
(55)
impedance spectroscopy
PEI-CNTs/GO
36 pmol/L
this work
amperometry technique
78 nmol/L
this work
potentiometry stripping
analysis
25 μmol/L
this work
Finally, the repeatability
of the proposed sensing platform was
investigated after 15 detection replicates of 0.5 mmol/L Pb(II) acetate
under the same working conditions using IS. All impedimetric responses
are shown in Figure S2, and all curves
obtained were identical with an acceptable standard deviation of 1.8%.
The results obtained are satisfactory in terms of repeatability, making
the proposed sensor platform attractive, with great potential to determine
trace levels of Pb ions.
Conclusions
CNTs
have been successfully functionalized with PEI through an
easy and environmentally friendly methodology. The idea consists of
wrapping CNTs with a PEI polymericchain through a sonication process.
It is a non-covalent approach to functionalize CNTs in a stable aqueous
solution easily. The attachment of PEI to CNTs wasconfirmed by UV–vis
and XPS spectroscopy. All graphene-based thin films were successfully
obtained by LbL films formed by the sequential addition of PEI-CNTs
and GO. The electrical characterizations show the resistive behavior
of the (PEI-CNT/GO)25 LbL film. On the other hand, we have
shown the ability of the LbL film as a sensing platform for the detection
of Pb(II) acetate.The detection mechanism associated with the
electrochemical techniques
studied with the functionalization of the microelectrodes allows us
to obtain a LOD of down to 36 pmol/L and a sensitivity up to 4.3 μAL/μmol,
with excellent repeatability. This is possible due to the high affinity
of heavy metal ions with the functional groups present in PEI-CNTs
and GO, allowing us to obtain a high-performance and sensitive detection
platform.
Experimental Section
Materials
and Composite Fabrication
The CNTs used here are multi-walled
CNTs having 95 wt % of purity,
lengths of 10–20 μm, and diameters of ∼100 nm,
acquired from CNT Co., Ltd., Seol, Korea. Poly(ethyleneimine) (50
wt % in water) solution and lead (II) acetate trihydrate were purchased
from Sigma-Aldrich. Graphene oxidewater dispersion with a concentration
of 0.4 wt % (4 mg/mL) was obtained from Graphenea Inc. A Pb(II) acetate
(0.01 M) stock solution was prepared with HCl (0.1 M) solution, and
Pb(II) acetate working solutions were prepared by successive dilutions
of the stock solution with HCl (0.1 M) solution. PEI-CNTs were prepared
by adding 10 mg of PEI in 10 mL of ultrapure water and sonicating
for 30 min. Then, 1 mg of CNTs were added to 10 mL of the PEI solution
(a concentration of 2 mg/mL) and sonicated for 4 h. The graphene oxide
dispersion was diluted in ultrapure water and sonicated for 30 min,
obtaining a stable GO sheetconcentration with a concentration of
1 mg/mL. For the LbL film formation, PEI-CNTs and GO solutions were
used as positive and negative polyelectrolytes, respectively. The
pH of each polyelectrolyte and wash water was adjusted to 3.4 using
0.1 mol/L HCl. The immersion times in the LbL assembly were fixed
at 10 min for each polyelectrolyte and 1 min for washing, thus producing
(PEI-CNT/GO) LbL films, n is the number
of deposited bilayers. All the solutions were prepared and diluted
by ultrapure water (18 MΩ × cm) acquired from a Sartorius
Arium Comfort system.(PEI-CNT/GO) LbL films were deposited onto gold IDEscontaining 60 fingers having
150 nm height (∼140 nm of Au onto a ∼10 nm Cr adhesive
layer), 3 mm length, and 40 μm width, with a finger separation
distance of 40 μm each other, patterned onto glass slides by
standard photolithography at the Brazilian Nanotechnology National
Laboratory (LNNano).
Characterization Techniques
A UV–vis
Biochrom Libra S60 spectrophotometer was used to record absorption
spectra from 190 to 800 nm. The chemical states of the constituent
elements of CNTs, PEI-CNTs, and Pb(II) acetate were investigated by
X-ray photoelectron spectroscopy (XPS) using a Thermo Alpha 110 hemispherical
analyzer. The XPS measurement was performed using an Al cathode (hν
= 1486.6 eV) as the X-ray source in the constant pass mode (CPM) using
20 eV pass energy. The XPS characterization of CNTs and lead (II)
powder acetate wascarried out on “as-received” materials,
and the PEI-CNT composite was drop-cast on the Si substrate. The collected
XPS spectra were fitted using Voigt line shapes and Shirley backgrounds,
simulated by the QUASE software. The morphology of the LbL films was
evaluated using a scanning electron microscope (Jeol JSM6490LV).The electrical characterizations of the LbL film were performed at
room temperature. I–V measurements
were carried out with a PGSTAT 302N potentiostat/galvanostat (Autolab)
between ±0.5 V using 50 mV/s. The impedance data were acquired
using a Solarton 1260A impedance analyzer coupled to a 1296 dielectric
interface with a sine-wave voltage signal amplitude of 25 mV, without
a dc bias in the frequency range 1 Hz to 1 MHz. The electrical characterizations
performed to detect Pb(II) acetate were carried out, keeping the solution
under stirring. IS, chronoamperometry techniques, and PSA were performed
with a PGSTAT 302N potentiostat/galvanostat (Autolab). IS and chronoamperometry
measurements were performed using the two contact pads of the IDEs.
However, for the PSA, an Ag/AgCl/KCl (0.1 M) reference electrode was
used. For the (PEI-CNT/GO)25 LbL film operation as an impedimetric
sensor, three cycles were carried out for each impedance measurement
from 1 MHz to 1 Hz, applying a 25 mVRMS AC voltage signal.
Moreover, aqueous solutions of 0.1 mol/L KCl, 7.4 μmol/L Pb(II)acetate in 0.1 mol/L HCl solution, and ultrapure water were used to
study the behavior of the electrical properties of the device and
its ability to distinguish between different analytes. The error bar
is the standard deviation of three repetitive measurements made at
each concentration value.
Authors: Bishun Khare; Patrick Wilhite; Benjamin Tran; Elico Teixeira; Kenneth Fresquez; Delphine Nna Mvondo; Charles Bauschlicher; M Meyyappan Journal: J Phys Chem B Date: 2005-12-15 Impact factor: 2.991
Authors: My Dung Jusselme; Franck Poly; Edouard Miambi; Philippe Mora; Manuel Blouin; Anne Pando; Corinne Rouland-Lefèvre Journal: Sci Total Environ Date: 2012-01-04 Impact factor: 7.963
Authors: Di Zhao; Jie Li; Chao Li; Albert L Juhasz; Kirk G Scheckel; Jun Luo; Hong-Bo Li; Lena Q Ma Journal: Environ Sci Technol Date: 2016-05-26 Impact factor: 9.028