Titanium dioxide and polypyrrole molecularly imprinted polymer nanocomposites based electrochemical sensor for highly selective detection of p-nonylphenol

Analytica Chimica Acta 1080 (2019) 84e94

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Titanium dioxide and polypyrrole molecularly imprinted polymer nanocomposites based electrochemical sensor for highly selective detection of p-nonylphenol Mingzhu Yu a, b, 1, Lina Wu b, 1, Jiaona Miao a, Wei Wei a, Anran Liu a, *, Songqin Liu a, ** a

Jiangsu Engineering Laboratory of Smart Carbon-Rich Materials and Device (CMD), Jiangsu Province Hi-Tech Key Laboratory for Bio-medical Research, School of Chemistry and Chemical Engineering, Southeast University, Nanjing, 211189, PR China Jiangsu Entry-exit Inspection and Quarantine Bureau Industrial Products Testing Center, PR China

b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


The electrochemical sensor was constructed by composition of TiO2 and PPy MIP.
PPy MIP showed specific recognition and strong absorption to pnonylphenol.
The electrochemical sensor exhibited high selectivity and sensitivity for the detection of p-nonylphenol.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 May 2019 Received in revised form 24 June 2019 Accepted 26 June 2019 Available online 28 June 2019

We developed a new electrochemical sensor based on TiO2 and polypyrrole (PPy) molecularly imprinted polymer (MIP) nanocomposites for the high selective detection of p-nonylphenol in food samples, which is considered as a kind of endocrine disrupting chemical and harmful to human health. With p-nonylphenol as template molecules, the molecularly imprinted polymer was synthesized by the chemical oxidative polymerization of pyrrole and deposited on the surface of TiO2 nanoparticles to form partially encapsulated PPy@TiO2 nanocomposites, denoted as NP-PPy@TiO2 MIP. p-Nonylphenol was bound in the PPy matrix through hydrogen bond and p-p interaction between p-nonylphenol and PPy skeleton. NPPPy@TiO2 MIP nanocomposites were modified onto glassy carbon electrode (GCE) and p-nonylphenol molecules were excluded from PPy layers by potentiostatic sweeping at the potential of 1.3 V. The asprepared electrochemical sensor obtained a large amount of micro cavities in PPy layer which could specially recognize and combine target molecules p-nonylphenol. After special adsorption of p-nonylphenol from samples, p-nonylphenol embedded in the PPy layer exhibited a strong differential pulse voltammetry (DPV) response at 0.56 V, which can be used for the detection of p-nonylphenol with a linearly proportional concentration range of 1.0  108 to 8  105 mol/L and a detection limit of 3.91  109 mol/L. The good stability, reproducibility and specificity of the resulting MIP electrochemical sensor are demonstrated. It might open a new window for investigation of selectively electrochemical sensing of small organic molecules from their analogues with the molecular imprinting technique. © 2019 Elsevier B.V. All rights reserved.

Keywords: Molecularly imprinted polymer Electrochemical sensor Polypyrrole p-Nonylphenol

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (A. Liu), [email protected] (S. Liu). 1 Mingzhu Yu and Lina Wu contributed equally to the work. https://doi.org/10.1016/j.aca.2019.06.053 0003-2670/© 2019 Elsevier B.V. All rights reserved.

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

1. Introduction p-Nonylphenol is a kind of alkylphenolic environmental hormone, which has been widely used in the production of surfactant, plastics, lubricant additives, antioxidant, textiles and other agricultural chemical products [1e4]. p-Nonylphenol is known for its estrogenic effects as a kind of endocrine disrupting chemical which can interfere with the secretion of estrogen in the human body at the concentrations present in the environment (from pM to nM) and hence can damage the reproductive system [5e7]. Therefore, it is important and urgent to develop a rapid, ultrasensitive and accurate method for the detection of p-nonylphenol residue in aqueous environment and food. Despite a number of advanced instruments have been developed for the determination of trace organic compounds in diverse samples, the detection of p-nonylphenol in food samples at low concentration level is still a challenge. Due to the complex substance of food samples, analytical methods for the determination of p-nonylphenol rely on high performance liquid chromatography (HPLC) [8] and gas chromatography (GC) for separation. Mass spectrometry(MS) [9], gas chromatography-mass spectrometry (GC-MS) [10], liquid chromatography-mass spectrometry (LC-MS) [11], high performance liquid chromatography-ultraviolet (HPLC-UV) [12], high performance liquid chromatography-fluorescence (HPLC-FL) [13] and other combined technologies have been utilized in p-nonylphenol determination. However, these methods usually require complex equipment and time-consuming sample pretreatment and extraction procedures [14,15]. The complexity of food sample substance results in the difficulty for the extraction of p-nonylphenol from sample matrix, furtherly reducing the accuracy for p-nonylphenol detection. Electrochemical analysis methods and electrochemical sensors have been adopted in recent years for their higher sensitivity, fast response, low cost, easy operation and realtime response [16]. However, it is also difficult for the electrochemical analysis methods to distinguish p-nonylphenol from its analogues such as octylphenol and other alkylphenols. To enhance the selectivity of electrochemical sensors, surface molecularly imprinted techniques could be feasible tools in the design and construction of electrochemical sensors [17e20]. Usually, for the formation of surface molecular imprinted polymer (SMIP), a monomer is polymerized onto the surface of carriers in the presence of the target molecules as templates [21,22]. Subsequently, these template molecules are removed from the polymer matrix, leaving a large amount of cavities with special and fixed shape and size tailored for further molecular recognition [23e25]. The binding sites are located in the thin layer of surface imprinted polymers, which can greatly reduce the “embedding” phenomenon and accelerate the recognition kinetics [26e28]. Therefore, it can theoretically be used for extraction and enrichment of the target analyte with high selectivity and efficiency [29,30]. Thus, molecularly imprinted polymer (MIP) electrochemical sensors combining the molecularly imprinted techniques with electrochemical sensors have been widely used in the detection of different organic pollutants, such as environmental hormone and pesticides [31e34]. The MIP electrochemical sensors could show great sensitivity, selectivity and stability [35,36]. Herein, we developed a novel electrochemical sensor based on titanium dioxide and polypyrrole molecularly imprinted polymer nanocomposites (PPy@TiO2 MIP/Nafion/GCE sensor) [37e40]. The electrochemical sensor was utilized for the rapid and ultrasensitive detection of p-nonylphenol with high selectivity. In this work, TiO2 nanoparticles were used as the carriers due to their large specific surface area [41e43]. TiO2 nanoparticles can improve the adsorption of the target molecules [26,44e46] and shorten the electrochemical response time which are benefit for the enhancement of

85

electrochemical response and the extension of detection range [47]. PPy is considered as an important conjugated polymer which exhibits controlled electric conductivity and good stability over a wide pH range [48e50]. Hydrogen bonds could form between PPy and p-nonylphenol. This interaction theoretically determines the high specificity for molecular recognition and high affinity of binding with target p-nonylphenol molecules [51,52]. NP-PPy@TiO2 MIP nanocomposites were modified onto the surface of the glassy carbon electrode (GCE) and the template molecules p-nonylphenol were removed by electrochemical oxidation at þ1.3 V in 0.2 M Na2HPO4 solution. The DPV response of the p-nonylphenol absorbed in PPy@TiO2 MIP/Nafion/GCE sensor was detected. To test the performance of the electrochemical sensor, samples with different concentrations of p-nonylphenol were introduced and the DPV responses were recorded. This electrochemical sensor exhibited high sensitivity, stability, reproducibility and specificity. 2. Experiment section 2.1. Chemicals and materials FeCl3
6H2O, K3[Fe(CN)6], K4[Fe(CN)6], KCl, NaOH, HCl, NaH2PO4$2H2O and Na2HPO4
12H2O were purchased from Sinopharm chemical Reagent Co., Ltd (Shanghai, China), pyrrole, titania dioxide nanoparticles (anatase, hydrophilic oleophilic bi-functional), bisphenol A(BPA), phenol and 4-bromophenol were purchased from Aladdin Reagent Company. p-Nonylphenol was purchased from Dr.E(Germany). Octylphenol(OP) and 4-aminophenol(p-AP) were purchased from J&K Chemicals. The milk powder (Nestle) was brought from the local supermarket Nanjing, China. 2.2. Characterization and measurements The morphologies of the as-prepared materials were studied by using a field-emission scanning electron microscope (SEM, Nova NanoSEM 450, FEI, USA). Transmission electron microscopy (TEM) was recorded on JEM2100 transmission electron microscopy (JEOL, Japan). Fourier transform infrared spectrum (FT-IR) was carried out with a Nicolet 4700 FT-IR Spectrometer (Thermo, USA) equipped with an attenuated total reflection setup. Raman spectra were recorded on an Invia Raman spectrometer (Renishaw) with a 514 nm laser. The surface area and pore size distribution were calculated by Brunauer-Emmett-Teller (BET) method according to nitrogen adsorption-desorption isotherms conducted on a NoveWin 1000e instrument (Quantachrome, USA). 2.3. The preparation of NP-PPy@TiO2 MIP 1.36 mmol pyrrole and 3 mL isopropanol solution containing 0.34 mmol nonylphenol were mixed and stirred for 1 h at room temperature for the pre-assembly of monomers and templates. 0.150 g TiO2 nanoparticles were dispersed in 30 mL aqueous solution of isopropanol (4 mL) and 2.4 M hydrochloride acid (26 mL) in a 50 mL jacketed pilot plant reactor. Then, 0.69 mmol FeCl3
6 H2O was added into the TiO2 dispersion at 0  C, followed by 30-min stirring until the solution turned pale yellow. The polymerization of pyrrole was initiated by adding pyrrole and nonylphenol mixture into the TiO2 dispersion containing FeCl3. The mixture was stirred for 5 h and the solution became black. Then, the resultant solution was centrifuged at 9000 rpm for 5 min to collect the black precipitate. The precipitate was washed by distilled water for three times and dried at 60  C overnight. The dried sample was ground into powder and stored in a 4  C refrigerator, which was denoted as NPPPy@TiO2 MIP. For comparison, the nonimprinted PPy@TiO2

86

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

nanocomposites were prepared using the same procedure mentioned above but without the addition of p-nonylphenol, denoted as PPy@TiO2 NIP. The p-nonylphenol-imprinted PPy nanocomposites were prepared without the addition of the carrier TiO2, and denoted as NP-PPy MIP. 2.4. Preparation of MIP-based electrochemical sensor The MIP-based electrochemical sensor was fabricated as follows: the bare glassy carbon electrode (GCE) with a diameter of 3 mm was polished with 0.05 mm alumina slurry, followed by ultrasonication in deionized water. 10 mL NP-PPy@TiO2 MIP suspension (5 mg/mL) was carefully dropped onto the cleaned GCE surface and dried at room temperature. Then, 5 mL Nafion aqueous solution(0.05 wt%) was coated onto the surface of GCE and dried thoroughly in the air. The removal process of template molecules was carried out by sweeping i-t curve at the positive potential of 1.3 V for 600 s every time according to the literature [53]. This process was performed on a standard three-electrode system on CHI 660D electrochemical analyzer (CH Instruments, USA). The NP-PPy@TiO2 MIP modified electrode was used as the working electrode in 0.1 M PBS solution (pH 6.0). Pt wire was used as the counter electrode and commercially saturated calomel electrode (SCE) was used as the reference electrode. Thus, the template molecules in NP-PPy@TiO2 MIP were completely removed from PPy matrix and the MIP-based electrochemical sensor (PPy@TiO2 MIP/Nafion/GCE sensor) with large amounts of molecular binding sites was obtained. For control experiments, the PPy@TiO2 NIP/Nafion/GCE sensor and NP-PPy MIP/ Nafion/GCE sensor were obtained using the same procedure mentioned above with the PPy@TiO2 NIP and NP-PPy MIP nanocomposites as the modified material respectively. The Cyclic voltammetry (CV) and Differential pulse voltammetry (DPV) for the characterization of electrochemical sensors were performed on a standard three-electrode system on CHI 660D electrochemical analyzer (CH Instruments, USA). The electrochemical impedance spectroscopy (EIS) measurements of the electrochemical sensors were performed on a standard threeelectrode system on Gamry Reference 600 potentiostat/galvanostat/ZRA (Gamry, U.S.A). Different modified electrodes were separately used as the working electrode. Pt wire was used as the counter electrode and commercially saturated calomel electrode (SCE) was used as the reference electrode. 2.5. Electrochemical detection of nonylphenol Prior to measurement, the PPy@TiO2 MIP/Nafion/GCE sensor was incubated in the mixed solution of isopropanol and PBS solution (pH 5.0) (2/13,v/v) containing different concentrations of pnonylphenol under stirring. Thereafter, the DPV measurement was performed in the scan range from 0.3 to 0.8 V at a scan rate of 100 mV s1. The pulse amplitude, pulse width, pulse period and quiet time were 50 mV, 0.05 s, 0.1 s and 2 s, respectively. All the experiments were conducted at room temperature. 2.6. Real-sample analysis 30 mg milk powder was added into a 45 mL mixed solution of isopropanol and PBS solution (pH 5.0)(2/13,v/v). After 30 min ultrasonication and 30 min stirring, the supernatant was filtrated through a sterile Millipore membrane (0.45 mm) and the filtrate was collected. Every 10 mL as-prepared milk powder solution was spiked with 0.22, 0.66 and 1.1 mg target p-nonylphenol respectively, then the results were analyzed using the proposed method.

3. Results and discussion 3.1. Morphology and structure of the PPy@TiO2 MIP and ControlPPy@TiO2 NIP The synthesis of NP-PPy@TiO2 MIP is illustrated in Scheme 1. Firstly, a precursor between the p-nonylphenol and pyrrole was formed due to the hydrogen-bonding and electrostatic interactions, which was necessary for the formation of recognition sites. Then, the molecularly imprinted PPy was synthesized by the chemical oxidative polymerization of pyrrole with FeCl3, and deposited on the surface of TiO2 nanoparticles to form partially encapsulated PPy@TiO2 nanocomposites, owing to the bond interaction, electrostatic interaction and sorption interaction at the interface between PPy and TiO2 nanoparticles. Finally, the NP-PPy@TiO2 MIP nanocomposites were modified onto the surface of the GCE and the template molecules were removed by potentiostatic sweeping at the potential of 1.3 V, leaving recognition sites available. When target molecules interacted with the PPy@TiO2 MIP modified electrode, the target molecules could enter into the cavities successfully due to the similarity in size and shape to the cavities. The embedded target molecules were electrochemical active and exhibited a strong DPV response due to the oxidation of target molecules. This electrochemical response could be utilized for the selective detection of p-nonylphenol and the possible electrochemical oxidation of the p-NP was listed as follows:

The morphology of the TiO2 nanoparticles, PPy@TiO2 MIP and PPy@TiO2 NIP are clearly displayed in Fig. 1. Due to the hydrophobic interaction, TiO2 nanoparticles aggregated into clusters. The size of TiO2 nanoparticles were about 20 nm. After polymerization, the surface of TiO2 nanoparticles was coated with PPy layers. The structures of MIP and NIP were investigated by FTIR and Raman spectra. As shown in FTIR spectra (Fig. 2A), the p-nonylphenol sample represented a typical peak located at 3410 cm1 and two double peaks ranging from 3000 to 2800 cm1, corresponding to the OeH and CeH stretching vibration, respectively. These characteristic peaks of p-nonylphenol could be observed easily in the FTIR spectra of NP-PPy@TiO2 MIP, while these peaks were not distinct in FTIR spectra of PPy@TiO2 NIP. This result indicated the successful incorporation of p-nonylphenol into the PPy matrix of NP-PPy@TiO2 MIP composites. As shown in Raman spectra (Fig. 2B), two prominent bands could be observed in the Raman spectra of the p-nonylphenol sample. The broad bands at 1580 cm1 and 1380 cm1 were assigned to C]C backbone stretching and the ring-stretching mode of benzene ring, respectively. In the Raman spectrum of PPy@TiO2 NIP, the 1580 cm1 band is assigned to C]C backbone stretching of PPy. The double bands at 1370 and 1330 cm1 are attributed to the ring-stretching mode of PPy. The presence of TiO2 did not influence the band from 2000 to 1000 cm1 for TiO2@PPy composites. For NP-PPy@TiO2 MIP, the incorporation of p-nonylphenol could raise the band intensity at 1380 cm1, which was demonstrated by the Raman spectra of NPPPy@TiO2 MIP. The band intensity ratio I1380/I1580 of NP-PPy@TiO2 MIP, PPy@TiO2 NIP and p-nonylphenol was 1.086, 0.975 and 1.599, respectively. The band intensity ratio I1380/I1580 of NP-PPy@TiO2 MIP was obviously higher than that of PPy@TiO2 NIP, due to the incorporation of p-nonylphenol. Both the FTIR and Raman spectra results imply the successful synthesis of NP-PPy@TiO2 MIP composites and the presence of p-nonylphenol in the PPy matrix. The formation of template binding sites could induce a large

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

87

Scheme 1. Preparation of PPy@TiO2 MIP and fabrication of PPy@TiO2 MIP/Nafion/GCE electrochemical sensor.

Fig. 1. SEM images of (a) TiO2, (b) NP-PPy@TiO2 MIP, (c) PPy@TiO2 NIP and TEM images of (d) TiO2, (e) NP-PPy@TiO2 MIP and (f) PPy@TiO2 NIP.

amount of micro cavities and thus change the pore structure of PPy@TiO2 nanocomposites. N2 adsorptionedesorption isotherms and pore size distributions of TiO2 nanoparticles and PPy@TiO2 MIP were investigated. As shown in Fig. 3A, the characterization results indicated that the TiO2 nanoparticles possessed a BET specific surface area of 384.30 m2 g1. The pore size distributions calculated by using the BJH method from absorption branches showed that the TiO2 nanoparticles had a sharp pore size distribution at 4.367 nm (Fig. 3B), which was owing to porous structure of TiO2 nanoparticle aggregate. After surface molecularly imprinting of PPy and following removal of templates, the specific surface area of

PPy@TiO2 MIP decreased to 180.74 m2 g1 (Fig. 3C). During the surface molecularly imprinting process, pyrrole was in-situ polymerized on the surface of TiO2 nanoparticles which filled the pores in TiO2 nanoparticle aggregation. The PPy coating on the surface of TiO2 caused that the amount of the pores with size of 4.367 nm reduced greatly. Compared with TiO2 nanoparticles, the additional pore size distribution peak at 1.193 nm could be observed in the pore size distribution curve of PPy@TiO2 MIP (Fig. 3D). These pores with size of 1.193 nm were corresponding to cavities formed after the removal of p-nonylphenol templates, which demonstrated the successful synthesis of molecular imprinted polymer.

88

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

Fig. 2. (A) FT-IR spectra and (B) Raman spectra of (a) TiO2, (b) p-nonylphenol, (c) NP-PPy@TiO2 MIP and (d) PPy@TiO2 NIP.

Fig. 3. The nitrogen adsorption-desorption isotherms of the (A) TiO2 nanoparticles and (B) PPy@TiO2 MIP at 298 K and corresponding pore size distribution of (C) TiO2 nanoparticles and (D) PPy@TiO2 MIP respectively.

3.2. Electrochemical characterization The conventional methods for template removal are using organic reagents or buffer solution as eluent. However, it is time

consuming and the template cannot be removed entirely. In this work, a simple method, potentiostatic sweeping was presented to exclude nonylphenol molecules from the PPy matrix until there was no obvious signal of p-nonylphenol. From the sweeping i-t

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

curve (shown in Fig. S1), a differential pulse voltammetry (DPV) oxidation peak at 0.56 V could be observed, which was attributed to the oxidation of p-nonylphenol embedded in the PPy layer. The current intensity could reflect the amount of -nonylphenol inside the PPy matrix. After sweeping for 600 s, the oxidation peak of pnonylphenol disappeared, indicating the complete removal of pnonylphenol from PPy layer on the PPy@TiO2 MIP modified electrode. The reason for the effective extraction of template is that the volume of PPy matrix expanded in Na2HPO4 solution as a result of anion ingress when the PPy skeleton became oxidized and charge positive under the applied voltage [54,55]. Template molecules could be easily excluded from the enlarged cavities. As shown in Fig. 4A, the construction of the modified electrodes was characterized by cyclic voltammetry (CV) by using 0.1 mol/L KCl containing 0.05 mol/L K3[Fe(CN)6] at 50 mV s1 in the range of 0.3 Ve0.7 V vs SCE. In curve a, well defined reversible redox peaks can be observed on the bare GCE. When the GCE was modified with PPy@TiO2 MIP, the peak current intensity enhanced obviously (curve b, Fig. 4A). This result suggested that the modification of PPy@TiO2 MIP could result in a larger electrochemical surface area, due to the cavities in the PPy matrix which could accelerate electron transfer of [Fe(CN)6]3-/4-. After incubated with p-nonylphenol, the MIP absorbed p-nonylphenol molecules and blocked the cavities in the PPy matrix. Thus, the redox peak current intensity decreased as the result of the limitation of electron transfer(curve c, Fig. 4A). In contrast, the electrode modified with PPy@TiO2 NIP (curve d, Fig. 4A) exhibited a lower peak current intensity than PPy@TiO2 MIP. As shown in Fig. 4B, the Nyquist plots of each modified electrode reflected the interfacial properties of the electrodes. The diameter of the semicircle equals the electron-transfer resistance (Ret) of the redox probe which reflects the conductivity of electrode materials and the restricted ion diffusion rate of the redox probe during the electrochemical process. Bare GCE exhibited the lowest Ret (curve a, Fig. 4B). Once modified with MIP or NIP, PPy@TiO2 hindered electron transfer. Thus, the NP-PPy@TiO2 MIP exhibited a larger semicircle (curve b, Fig. 4B). After removal of p-nonylphenol templates, the Ret decreased (curve c, Fig. 4B). The reason is that the release of templates increased the amount of the cavities where p-nonylphenol templates located and formed extra ion diffusing channels, thus accelerating the ion diffusion at the liquid-solid interface.

89

When the PPy@TiO2 MIP absorbed p-nonylphenol in sample solution, p-nonylphenol embedded with PPy@TiO2 MIP and filled the cavities (curve d, Fig. 4B). Compared with PPy@TiO2 MIP, PPy@TiO2 NIP (curve e, Fig. 4B) showed a larger Ret, which was attributed to the non-porous structure of the PPy matrix.

3.3. Electrochemical behavior of the modified sensor Fig. 5A shows the DPV responses of the PPy@TiO2 MIP/Nafion/ GCE in PBS (pH 7.0). There is no DPV oxidation peak for p-nonylphenol (curve a, Fig. 5A), implying the complete removal of pnonylphenol inside the PPy. After incubating PPy@TiO2 MIP/Nafion/ GCE sensor in 1 mM (curve b, Fig. 5A), 10 mM (curve c, Fig. 5A) and 100 mM (curve d, Fig. 5A) p-nonylphenol solution, the DPV oxidation peak for p-nonylphenol appeared again, and the current response increased with the increase of the concentration of pnonylphenol solution for adsorption. Since the DPV measurement was performed in p-nonylphenol-free solution, it indicated that the DPV response was caused only by the redox of p-nonylphenol embedded inside the PPy film. The PPy@TiO2 MIP/Nafion/GCE sensor exhibited specific molecular recognition to target p-nonylphenol and low nonspecific adsorption of p-nonylphenol and other organic molecules. The nonspecific adsorption of the PPy@TiO2 MIP/Nafion/GCE sensor was mainly due to the interaction between organic molecules and polypyrrole such as hydrophobic interaction. The specificity of molecule recognition between PPy@TiO2 MIP and p-nonylphenol was verified by comparing the DPV responses of different modified electrodes after incubated in 10 mM p-nonylphenol solution. As shown in Fig. 5B, PPy@TiO2 MIP/Nafion/GCE exhibited the strongest DPV response (curve a, Fig. 5B), due to the specific molecule recognition and binding between MIP and p-nonylphenol. p-Nonylphenol could enter into the cavities successfully due to the similarity in size and shape to the cavities, resulting in the effective absorption of MIP. PPy MIP/Nafion/GCE modified electrode showed a weaker DPV response (curve b, Fig. 5B) than PPy@TiO2 MIP/ Nafion/GCE. The reason is that the TiO2/PPy core-shell structure increased the surface area of the MIP nanocomposites and improved the absorption efficiency. The TiO2/Nafion/GCE modified electrode (curve c, Fig. 5B) and PPy@TiO2 NIP/Nafion/GCE modified electrode (curve d, Fig. 5B) had rather weak DPV responses. The

Fig. 4. (A) CV responses of 0.05 M K3[Fe(CN)6] solution containing 0.1 M KCl at (a) bare GCE sensor, (b) PPy@TiO2 MIP/Nafion/GCE sensor, (c) PPy@TiO2 MIP/Nafion/GCE sensor incubating in 0.1 mM p-nonylphenol (d) PPy@TiO2 NIP/Nafion/GCE sensor, and (B) EIS responses of 0.05 M K3[Fe(CN)6] solution containing 0.1 M KCl at (a) bare GCE sensor, (b) NPPPy@TiO2 MIP/Nafion/GCE sensor, (c) PPy@TiO2 MIP/Nafion/GCE sensor, (d) PPy@TiO2 MIP/Nafion/GCE sensor incubating in 0.1 mM p-nonylphenol, (e) PPy@TiO2 NIP/Nafion/GCE sensor.

90

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

Fig. 5. (A) DPV responses of 0.1 M PBS solution at (a) PPy@TiO2 MIP/Nafion/GCE sensor, MIP sensor incubating in (b) 1 mM, (c) 10 mM and (d)100 mM p-nonylphenol solution. (B) DPV responses of 0.1 M PBS solution (pH 6.0) at (a) PPy@TiO2 MIP/Nafion/GCE sensor, (b) PPy MIP/Nafion/GCE modified electrode, (c) TiO2/Nafion/GCE modified electrode, (d) PPy@TiO2 NIP/Nafion/GCE modified electrode after incubated in 10 mM p-nonylphenol solution, respectively.

DPV response at TiO2/Nafion/GCE and NIP/Nafion/GCE modified electrode was owing to the oxidation of p-nonylphenol nonspecifically adsorbed on the surface of TiO2 nanoparticles and PPy@TiO2 NIP. For further study of nonspecific adsorption of the proposed MIP sensor, the DPV responses caused by nonspecific adsorption of PPy@TiO2 NIP were measured. As shown in Fig. S2, DPV responses at NIP/Nafion/GCE electrode and PPy@TiO2 MIP/ Nafion/GCE sensor incubating in p-nonylphenol solution of different concentration for 20 min were recorded and compared. The DPV response at NIP/Nafion/GCE modified electrode was much weaker than MIP/Nafion/GCE sensor and increased slowly with the increase of the concentration of p-nonylphenol solutions. All the test results demonstrated the specificity of molecule recognition between PPy@TiO2 MIP and p-nonylphenol and low nonspecific adsorption of the MIP sensor. 3.4. Optimization of MIP synthesis and detecting conditions The main parameters that can crucially affect the formation of the PPy@TiO2 MIP are the ratio of pyrrole and p-nonylphenol, ratio of pyrrole to TiO2, incubation time and pH of the detection. The ratio of monomer and template in the polymerization process will affect the amount of the imprinted sites in the polymer matrix, which will further affect the electrochemical behavior of the sensor. In order to obtain an optimal recognition film, the effect of pyrrole to p-nonylphenol molar ratio on the current response of PPy@TiO2 MIP/Nafion/GCE was investigated in PBS (pH 7.0) containing 10 mM p-nonylphenol, and the obtained results were shown in Fig. S3A. The largest current response was obtained at pyrrole to p-nonylphenol molar ratio of 4. While the pyrrole to pnonylphenol molar ratio was higher than 4, the current response decreased, probably because the amount of the template molecules decreased, the formation of recognition sites or binding cavities in the PPy film decreased. Similarly, the pyrrole to pnonylphenol molar ratio below 4 also resulted in a decrease in current response. The possible reason was that the amount of monomer was too little to combine enough template molecules, causing fewer number of recognition sites in the PPy film. Therefore, the optimal pyrrole to p-nonylphenol molar ratio for the polymerization was found to be 4.

The ratio of monomer and carrier (TiO2) in the polymerization process will also influence the electrochemical behavior of the sensor. In order to obtain an optimal recognition film, the effect of pyrrole to TiO2 mass ratio on the current response of PPy@TiO2 MIP/ Nafion/GCE sensor was investigated in PBS (pH 7.0) containing 20 mM p-nonylphenol, and the obtained results were shown in Fig. S3B. The largest current response was obtained at pyrrole to TiO2 mass ratio of 0.6. When pyrrole to TiO2 mass ratio was below 0.6, the current response would increase as the increased pyrrole to TiO2 mass ratio because it could be beneficial to promoting the amount of binding sites. With the continuous increase of pyrrole to TiO2 mass ratio above 0.6, self-polymerization of pyrrole would tend to occur instead of polymerization onto the surface of TiO2. Incubating time is another key factor which affects the sensitivity. As shown in Fig. S3C, the current response increased with the increase of the incubating time and reached a stable value at 20 min. With the extension of time above 20 min, the current response changed slightly, since the absorption reached the equilibrium state. Therefore, the optimal incubating time is 20 min. The detection sensitivity is largely determined by the pH of the detection solution. As shown in Fig. S3D, the current response gained the highest value with the pH of 6.0. The higher pH of the solution could induce the formation of phenoxy anion from pnonylphenol more easily. However, when the solution was neutral or basic, the deprotonation effect could result in undoping of PPy and destroyed the conjugated structure, further hindering the electron transfer in PPy skeleton and decreasing the conductivity of PPy. Thus, the optimized detecting pH should be 6.0. 3.5. Quantitative determination of p-nonylphenol Upon the optimum conditions, the DPV responses varied with different concentration of target p-nonylphenol (Fig. 6A), and it can be found that current intensity is related to the concentration of pnonylphenol. Accordingly, the current intensity was proportional to concentration of target in two linear ranges: 1.00  108 to 1.00  106 mol/L and 1.00  106 to 8.00  105 mol/L (Fig. 6B). The linear regression equation was i1(mA) ¼ 2.500 c1(mM) þ 1.223 (R2 ¼ 0.997) for the first linear range and i2(mA) ¼ 0.126 c2(mM) þ 3.7157 (R2 ¼ 0.998) for the second linear range. The limit

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

91

Fig. 6. DPV responses at PPy@TiO2 MIP/Nafion/GCE sensor incubating in 8.00  105, 6.00  105, 4.00  105, 2.00  105, 1.00  105, 1.00  106, 8.00  107, 5.00  107, 1.00  107, 5.00  108, 1.00  108 mol/L p-nonylphenol solution (from top to bottom) for 20 min. (B) Linear calibration curves for PPy@TiO2 MIP. All DPV currents were recorded in 0.1 M PBS solution (pH 6.0).

of detection is 3.91  109 mol/L (LOD, S/N ¼ 3), which was much lower than the reported ELISA (94 ng/mL), LCMS/MS (47 ng/mL) [11] or other methods shown in Table 1. The existence of two linear ranges might be attributed to the affinity between the imprinted sites and p-nonylphenol. When the p-nonylphenol concentration is lower, the molecules tend to prefer the cavities with high affinity cavities located in the upper part of the film. Conversely, the lower affinity sites located more deeply are occupied when the p-nonylphenol concentration is higher, causing a decline in the linear slope.

3.6. Stability, reproducibility and specificity The selectivity of the proposed PPy@TiO2 MIP/Nafion/GCE sensor towards p-nonylphenol and its analogues was investigated by measuring the DPV responses to 4-octylphenol(OP), bisphenol A(BPA), 4-aminophenol(p-AP), phenol and 4-bromophenol at a concentration of 1.0  105 M that was the same as p-nonylphenol, As shown in Fig. 7, OP, BPA, p-AP, phenol and 4-bromophenol had similar structure to p-nonylphenol, but their DPV responses were 5 times lower than that of p-nonylphenol. The result indicated that

Table 1 Comparison of Different Sensors for the Determination of p-Nonylphenol. Methods

Linear range(M)

LOD(M)

References

AuNPs/PILs MIP/TiO2-AuNPs/GCE graphene/DNA/GCE nanopyramid BDD MIP/nitrogen-doped graphene nanoribbons/GCE PPy@TiO2 MIP/Nafion/GCE

1  106 e 1.2  104 9.5  107 e 4.8  104 5  108 e 4  106 1  109 e 1  107 4  108 e 6  107 1  108 e 1  106 1  106 e 8  105

3.3  108 9.5  107 1  108 2.3  1010 8  109 3.91  109

[56] [30] [57] [58] [59] this work

Fig. 7. (A) Selectivity of PPy@TiO2 MIP/Nafion/GCE sensor and PPy@TiO2 NIP/Nafion/GCE modified electrode to p-nonylphenol and the interferences. The concentrations of pnonylphenol and the interferences were all 10 mM. (B) Chemical structures of p-nonylphenol and the interferences.

92

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

Fig. 8. Stability test of the molecularly imprinted polymer electrochemical sensor via DPV responses of 10 mM p-nonylphenol at the PPy@TiO2 MIP/Nafion/GCE sensor in 0.1 M PBS (pH ¼ 6.0) after one week, two weeks, three weeks and four weeks respectively.

Table 2 Recovery Results of p-Nonylphenol Added in Real Samples. Samples

Found (mM)

Nonylphenol added (mM)

Total found (mM)

Peak current (mA)

Recoveries (%)

RSD (%)

Nestle milk powder

ND

0.000 0.100 0.300 0.500

0 0.099 0.302 0.492

0 1.471 1.978 2.453

\ 99.0 100.7 98.4

\ 3.42 4.80 6.35

DPV response of the PPy@TiO2 MIP/Nafion/GCE sensor was mainly caused by a specific molecular recognition and binding between pnonylphenol and its molecularly imprinted polymer layers. The storage stability of the proposed PPy@TiO2 MIP/Nafion/GCE electrochemical sensor was investigated by storing the same PPy@TiO2 MIP at 4  C for different periods before the DPV detection of 1.0  106 M p-nonylphenol. It was clear that 92.22% of the DPV response (Fig. 8) for the sensor was retained after storage for 4 weeks, indicating that the prepared PPy@TiO2 MIP/Nafion/GCE electrochemical sensor possesses good stability and potential for practical application. The stability of the PPy@TiO2 MIP/Nafion/GCE electrochemical sensor mainly relies on the binding affinity between the polymers and the template molecules. After a long time storage, the PPy@TiO2 MIP maintained its pore structure and showed good binding ability to the target. Attributed to the stable system, a good accuracy and reproducibility were also obtained (Table S1). Various concentrations of standard p-nonylphenol solution were detected five times respectively with the proposed method (in the way of intra-assay and inter-assay), the intra-assay variability (CV % ¼ SD/mean) of 2.75% and the inter-assay variability of 6.35% both suggest a good reproducibility of the proposed MIP electrochemical sensor. 3.7. Real sample analysis A significant and challenging factor for p-nonylphenol analysis is the applicability in food samples (Table 2). 10 mL milk powder solution samples spiked with 0.22, 0.66 and 1.1 mg p-nonylphenol respectively and were analyzed with the proposed method. Each concentration of the sample was tested 5 times. The results were shown in Table 2 and exhibited good recoveries varied in the range of 98.4%e100.7%. The relative standard derivations (RSD) were

from 3.42% to 6.35%, indicating that our method had high precision and good reliability for detection of p-nonylphenol in real examinations and held great potential for food detection. 4. Conclusion With the surface modification of PPy@TiO2 MIP on the GCE, a highly sensitive and selective electrochemical sensor was proposed for the determination of p-nonylphenol. The excellent linear relationship was acquired between the DPV current intensity and the p-nonylphenol concentration from 1.0  108 to 8  105 mol/L and LOD of this method was 3.91  109 mol/L. The proposed method exhibited a high selectivity which could distinguish the target pnonylphenol from its analogues such as octylphenol and bisphenol A in solution. This method was also demonstrated to be feasible in real sample analysis. This designed sensing strategy not only served as universal tool that can be expanded to detect different small organic molecule targets, but also provided a powerful tool for investigation of highly selective electrochemical sensing processes with numerous MIP based nanomaterials. Conflict of interest The authors declared that they have no conflicts of interest to this work. Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94

Notes The authors declare no competing financial interest. Acknowledgements This work was financially supported by the National Key Research and Development Program of China (No. 2018YFC1602800), the National Key Research and Development Program of China (No. 2016YFF0203703), the National Natural Science Foundation of China (21505018), Fundamental Research Funds for the Central Universities (2242017k30001). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.aca.2019.06.053. References [1] P.H. Brunner, S. Capri, A. Marcomini, W. Giger, Occurrence and behavior of linear alkylbenzenesulfonates, nonylphenol, nonylphenol monophenol and nonylphenol diethoxylates in sewage and sewage-sludge treatment, Water Res. 22 (1988) 1465e1472. [2] N.G. Coldham, S. Sivapathasundaram, M. Dave, L.A. Ashfield, T.G. Pottinger, C. Goodall, M.J. Sauer, Biotransformation, tissue distribution, and persistence of 4-nonylphenol residues in juvenile rainbow trout (Oncorhynchus mykiss), Drug Metab. Dispos. 26 (1998) 347e354. [3] J.B. Colerangle, D. Roy, Exposure of environmental estrogenic compound nonylphenol to noble rats alters cell-cycle kinetics in the mammary gland, Endocrine 4 (1996) 115e122. [4] A. Soares, B. Guieysse, B. Jefferson, E. Cartmell, J.N. Lester, Nonylphenol in the environment: a critical review on occurrence, fate, toxicity and treatment in wastewaters, Environ. Int. 34 (2008) 1033e1049. [5] J. Shan, B.Q. Jiang, B. Yu, C.L. Li, Y.Y. Sun, H.Y. Guo, J.C. Wu, E. Klumpp, A. Schaffer, R. Ji, Isomer-specific degradation of branched and linear 4nonylphenol isomers in an oxic soil, Environ. Sci. Technol. 45 (2011) 8283e8289. [6] A.M. Soto, H. Justicia, J.W. Wray, C. Sonnenschein, Para-nonyl-phenol - an estrogenic xenobiotic released from modified polystyrene, Environ. Health Perspect. 92 (1991) 167e173. [7] K.E. Tollefsen, S. Elkvar, E.F. Finne, O. Fogelberg, I.K. Gregersen, Estrogenicity of alkylphenols and alkylated non-phenolics in a rainbow trout (Oncorhynchus mykiss) primary hepatocyte culture, Ecotoxicol. Environ. Saf. 71 (2008) 370e383. [8] X.Y. Liu, X.Y. Zhang, H.X. Zhang, M.C. Liu, A chemometric strategy for optimization of solid-phase microextraction: determination of bisphenol a and 4nonylphenol with HPLC, J. Chromatogr. Sci. 46 (2008) 596e600. [9] N. Fabregat-Cabello, A. Castillo, J.V. Sancho, F.V. Gonzalez, A.F. Roig-Navarro, Fast methodology for the reliable determination of nonylphenol in water samples by minimal labeling isotope dilution mass spectrometry, J. Chromatogr. A 1301 (2013) 19e26. [10] S.S. Luo, L. Fang, X.W. Wang, H.T. Liu, G.F. Ouyang, C.Y. Lan, T.G. Luan, Determination of octylphenol and nonylphenol in aqueous sample using simultaneous derivatization and dispersive liquid-liquid microextraction followed by gas chromatography-mass spectrometry, J. Chromatogr. A 1217 (2010) 6762e6768. [11] N. Salgueiro-Gonzalez, E. Concha-Grana, I. Turnes-Carou, S. Muniategui-Lorenzo, P. Lopez-Mahia, D. Prada-Rodriguez, Determination of alkylphenols and bisphenol A in seawater samples by dispersive liquid-liquid microextraction and liquid chromatography tandem mass spectrometry for compliance with environmental quality standards (Directive 2008/105/EC), J. Chromatogr. A 1223 (2012) 1e8. [12] H.K. Shih, T.Y. Shu, V.K. Ponnusamy, J.F. Jen, A novel fatty-acid-based in-tube dispersive liquid-liquid microextraction technique for the rapid determination of nonylphenol and 4-tert-octylphenol in aqueous samples using highperformance liquid chromatography-ultraviolet detection, Anal. Chim. Acta 854 (2015) 70e77. [13] Q. Zhou, Y. Gao, G. Xie, Determination of bisphenol A, 4-n-nonylphenol, and 4tert-octylphenol by temperature-controlled ionic liquid dispersive liquidphase microextraction combined with high performance liquid chromatography-fluorescence detector, Talanta 85 (2011) 1598e1602. [14] N. Salgueiro-Gonzalez, S. Castiglioni, E. Zuccato, I. Turnes-Carou, P. LopezMahia, S. Muniategui-Lorenzo, Recent advances in analytical methods for the determination of 4-alkylphenols and bisphenol A in solid environmental matrices: a critical review, Anal. Chim. Acta 1024 (2018) 39e51. [15] N. Salgueiro-Gonzalez, S. Muniategui-Lorenzo, P. Lopez-Mahia, D. PradaRodriguez, Trends in analytical methodologies for the determination of alkylphenols and bisphenol A in water samples, Anal. Chim. Acta 962 (2017)

93

1e14. [16] W.S. Zhou, B. Zhao, X.H. Huang, X.D. Yang, Electrochemical determination of 4-nonylphenol based on graphene-chitosan modified glassy carbon electrode, Chin. J. Anal. Chem. 41 (2013) 675e680. [17] X. Wang, J. Dong, H. Ming, S. Ai, Sensing of glycoprotein via a biomimetic sensor based on molecularly imprinted polymers and graphene-Au nanoparticles, Analyst 138 (2013) 1219e1225. [18] H. Wang, S. Yao, Y. Liu, S. Wei, J. Su, G. Hu, Molecularly imprinted electrochemical sensor based on Au nanoparticles in carboxylated multi-walled carbon nanotubes for sensitive determination of olaquindox in food and feedstuffs, Biosens. Bioelectron. 87 (2017) 417e421. [19] J. Yao, M. Chen, N. Li, C. Liu, M. Yang, Experimental and theoretical studies of a novel electrochemical sensor based on molecularly imprinted polymer and B, N, F-CQDs/AgNPs for enhanced specific identification and dual signal amplification in highly selective and ultra-trace bisphenol S determination in plastic products, Anal. Chim. Acta 1066 (2019) 36e48. [20] R.A.S. Couto, S.S. Costa, B. Mounssef Jr., J.G. Pacheco, E. Fernandes, F. Carvalho, C.M.P. Rodrigues, C. Delerue-Matos, A.A.C. Braga, L.M. Goncalves, M.B. Quinaz, Electrochemical sensing of ecstasy with electropolymerized molecularly imprinted poly(o-phenylenediamine) polymer on the surface of disposable screen-printed carbon electrodes, Sensor. Actuator. B Chem. 290 (2019) 378e386. [21] F. Deng, Y.X. Li, X.B. Luo, L.X. Yang, X.M. Tu, Preparation of conductive polypyrrole/TiO2 nanocomposite via surface molecular imprinting technique and its photocatalytic activity under simulated solar light irradiation, Colloids Surf., A 395 (2012) 183e189. [22] S. Li, L. Fang, M.M. Ye, Y. Zhang, Enhanced adsorption of norfloxacin on modified TiO2 particles prepared via surface molecular imprinting technique, Desalin.Water Treat. 57 (2016) 408e418. [23] Z. Liu, M. Jin, H. Lu, J. Yao, X. Wang, G. Zhou, L. Shui, Molecularly imprinted polymer decorated 3D-framework of functionalized multi-walled carbon nanotubes for ultrasensitive electrochemical sensing of Norfloxacin in pharmaceutical formulations and rat plasma, Sensor. Actuator. B Chem. 288 (2019) 363e372. [24] C.R. Rawool, A.K. Srivastava, A dual template imprinted polymer modified electrochemical sensor based on Cu metal organic framework/mesoporous carbon for highly sensitive and selective recognition of rifampicin and isoniazid, Sensor. Actuator. B Chem. 288 (2019) 493e506. [25] L. Li, H. Zheng, L. Guo, L. Qu, L. Yu, A sensitive and selective molecularly imprinted electrochemical sensor based on Pd-Cu bimetallic alloy functionalized graphene for detection of amaranth in soft drink, Talanta 197 (2019) 68e76. [26] M. Bosserdt, J. Erdossy, G. Lautner, J. Witt, K. Kohler, N. Gajovic-Eichelmann, A. Yarman, G. Wittstock, F.W. Scheller, R.E. Gyurcsanyi, Microelectrospotting as a new method for electrosynthesis of surface-imprinted polymer microarrays for protein recognition, Biosens. Bioelectron. 73 (2015) 123e129. [27] S.R. Carter, S. Rimmer, Surface molecularly imprinted polymer core-shell particles, Adv. Funct. Mater. 14 (2004) 553e561. [28] S.J. Hua, L. Zhao, L.Y. Cao, X.Q. Wang, J.S. Gao, C.M. Xu, Fabrication and evaluation of hollow surface molecularly imprinted polymer for rapid and selective adsorption of dibenzothiophene, Chem. Eng. J. 345 (2018) 414e424. [29] J.B. Ai, H. Guo, R. Xue, X. Wang, X. Lei, W. Yang, A self-probing, gate-controlled, molecularly imprinted electrochemical sensor for ultrasensitive determination of p-nonylphenol, Electrochem. Commun. 89 (2018) 1e5. [30] J. Huang, X. Zhang, S. Liu, Q. Lin, X. He, X. Xing, W. Lian, D. Tang, Development of molecularly imprinted electrochemical sensor with titanium oxide and gold nanomaterials enhanced technique for determination of 4-nonylphenol, Sensor. Actuator. B Chem. 152 (2011) 292e298. [31] D.N. Pei, A.Y. Zhang, X.Q. Pan, Y. Si, H.Q. Yu, Electrochemical sensing of bisphenol a on facet-tailored TiO2 single crystals engineered by inorganicframework molecular imprinting sites, Anal. Chem. 90 (2018) 3165e3173. [32] B. Wang, O.K. Okoth, K. Yan, J. Zhang, A highly selective electrochemical sensor for 4-chlorophenol determination based on molecularly imprinted polymer and PDDA-functionalized graphene, Sensor. Actuator. B Chem. 236 (2016) 294e303. [33] M.P. Sooraj, B. Mathew, Fabrication of a structure-specific molecular imprinted polymer-based electrochemical sensor based on CuNP-decorated vinylfunctionalized graphene for the detection of parathion methyl in vegetable and fruit samples, Food Anal. Methods 12 (2019) 1028e1039. [34] Y. Li, L. Zhang, Y. Dang, Z. Chen, R. Zhang, Y. Li, B.-C. Ye, A robust electrochemical sensing of molecularly imprinted polymer prepared by using bifunctional monomer and its application in detection of cypermethrin, Biosens. Bioelectron. 127 (2019) 207e214. [35] M.A. Hammam, H.A. Wagdy, R.M. El Nashar, Moxifloxacin hydrochloride electrochemical detection based on newly designed molecularly imprinted polymer, Sensor. Actuator. B Chem. 275 (2018) 127e136. [36] W. Zheng, Z. Xiong, H. Li, S. Yu, G. Li, L. Niu, W. Liu, Electrodeposited Pt@ Molecularly imprinted polymer core-shell nanostructure: enhanced sensing platform for sensitive and selective detection of bisphenol A, Sensor. Actuator. B Chem. 272 (2018) 655e661. [37] H. Mao, J.C. Liang, H.F. Zhang, Q. Pei, D.L. Liu, S.Y. Wu, Y. Zhang, X.M. Song, Poly(ionic liquids) functionalized polypyrrole/graphene oxide nanosheets for electrochemical sensor to detect dopamine in the presence of ascorbic acid, Biosens. Bioelectron. 70 (2015) 289e298. [38] T. Qian, C.F. Yu, X. Zhou, P.P. Ma, S.S. Wu, L.N. Xu, J. Shen, Ultrasensitive

94

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

M. Yu et al. / Analytica Chimica Acta 1080 (2019) 84e94 dopamine sensor based on novel molecularly imprinted polypyrrole coated carbon nanotubes, Biosens. Bioelectron. 58 (2014) 237e241. A. Turco, S. Corvaglia, E. Mazzotta, Electrochemical sensor for sulfadimethoxine based on molecularly imprinted polypyrrole: study of imprinting parameters, Biosens. Bioelectron. 63 (2015) 240e247. X.R. Xing, S. Liu, J.H. Yu, W.J. Lian, J.D. Huang, Electrochemical sensor based on molecularly imprinted film at polypyrrole-sulfonated graphene/hyaluronic acid-multiwalled carbon nanotubes modified electrode for determination of tryptamine, Biosens. Bioelectron. 31 (2012) 277e283. H.J. Shi, Y.L. Wang, C.J. Tang, W.K. Wang, M.C. Liu, G.H. Zhao, Mechanisim investigation on the enhanced and selective photoelectrochemical oxidation of atrazine on molecular imprinted mesoporous TiO2, Appl. Catal. B Environ. 246 (2019) 50e60. C.-T. Wu, P.-Y. Chen, J.-G. Chen, V. Suryanarayanan, K.-C. Ho, Detection of nicotine based on molecularly imprinted TiO2-modified electrodes, Anal. Chim. Acta 633 (2009) 119e126. L. Yang, H. Zhao, S. Fan, B. Li, C.-P. Li, A highly sensitive electrochemical sensor for simultaneous determination of hydroquinone and bisphenol A based on the ultrafine Pd nanoparticle@TiO2 functionalized SiC, Anal. Chim. Acta 852 (2014) 28e36. A. Bouarioua, M. Zerdaoui, Photocatalytic activities of TiO2 layers immobilized on glass substrates by dip-coating technique toward the decolorization of methyl orange as a model organic pollutant, J. Environ.Chem. Eng. 5 (2017) 1565e1574. H.R. Geng, S.S. Miao, S.F. Jin, H. Yang, A newly developed molecularly imprinted polymer on the surface of TiO2 for selective extraction of triazine herbicides residues in maize, water, and soil, Anal. Bioanal. Chem. 407 (2015) 8803e8812. N. Mizutani, D.-H. Yang, R. Selyanchyn, S. Korposh, S.-W. Lee, T. Kunitake, Remarkable enantioselectivity of molecularly imprinted TiO2 nano-thin films, Anal. Chim. Acta 694 (2011) 142e150. Z.J. Huang, Z.M. Zhang, Q. Xia, C.L. Li, Y.B. Yun, Surface molecularly imprinted polymer microspheres based on nano-TiO2 for selective recognition of kaempferol, J. Appl. Polym. Sci. 134 (2017). T. Qian, C.F. Yu, S.S. Wu, J. Shen, In situ polymerization of highly dispersed polypyrrole on reduced graphite oxide for dopamine detection, Biosens. Bioelectron. 50 (2013) 157e160. Y. Yang, G. Fang, X. Wang, M. Pan, H. Qian, H. Liu, S. Wang, Sensitive and selective electrochemical determination of quinoxaline-2-carboxylic acid

[50]

[51]

[52]

[53]

[54] [55]

[56]

[57]

[58]

[59]

based on bilayer of novel poly(pyrrole) functional composite using one-step electro-polymerization and molecularly imprinted poly(o-phenylenediamine), Anal. Chim. Acta 806 (2014) 136e143. A.D. Arulraj, M. Vijayan, V.S. Vasantha, Highly selective and sensitive simple sensor based on electrochemically treated nano polypyrrole-sodium dodecyl sulphate film for the detection of para-nitrophenol, Anal. Chim. Acta 899 (2015) 66e74. F. Gao, X.H. Hou, A. Wang, G.W. Chu, W. Wu, J.F. Chen, H.K. Zou, Preparation of polypyrrole/TiO2 nanocomposites with enhanced photocatalytic performance, Particuology 26 (2016) 73e78. X. Liu, X. Wang, F. Tan, H. Zhao, X. Quan, J. Chen, L. Li, An electrochemically enhanced solid-phase microextraction approach based on molecularly imprinted polypyrrole/multi-walled carbon nanotubes composite coating for selective extraction of fluoroquinolones in aqueous samples, Anal. Chim. Acta 727 (2012) 26e33. H. Shi, G. Zhao, M. Liu, Z. Zhu, A novel photoelectrochemical sensor based on molecularly imprinted polymer modified TiO2 nanotubes and its highly selective detection of 2,4-dichlorophenoxyacetic acid, Electrochem. Commun. 13 (2011) 1404e1407. E. Smela, Conjugated polymer actuators for biomedical applications, Adv. Mater. 15 (2003) 481e494. B.J. West, T.F, Otero, B. Shapiro, E. Smela, Chronoamperometric study of conformational relaxation in PPy(DBS), J. Phys. Chem. B 113 (2009) 1277e1293. J. Ren, J. Gu, L. Tao, M. Yao, X. Yang, W. Yang, A novel electrochemical sensor of 4-nonylphenol based on a poly(ionic liquid) hollow nanosphere/gold nanoparticle composite modified glassy carbon electrode, Anal. Methods 7 (2015) 8094e8099. L. Zeng, A. Zhang, X. Zhu, C. Zhang, Y. Liang, J. Nan, Electrochemical determination of nonylphenol using differential pulse voltammetry based on a graphene-DNA-modified glassy carbon electrode, J. Electroanal. Chem. 703 (2013) 153e157. X. Yuan, N. Gao, X. Gao, D. Qiu, R. Xu, Z. Sun, Z. Jiang, J. Liu, H. Li, Nanopyramid boron-doped diamond electrode realizing nanomolar detection limit of 4nonylphenol, Sensor. Actuator. B Chem. 281 (2019) 830e836. Y. Pan, L. Shang, F. Zhao, B. Zeng, A novel electrochemical 4-nonyl-phenol sensor based on molecularly imprinted poly (o-phenylenediamine-co-o-toluidine)-nitrogen-doped graphene nanoribbons-ionic liquid composite film, Electrochim. Acta 151 (2015) 423e428.