In-site synthesis molecular imprinting Nb2O5 –based photoelectrochemical sensor for bisphenol A detection

Author’s Accepted Manuscript In-site synthesis molecular imprinting Nb2O5 – based photoelectrochemical sensor for bisphenol A detection Pan Gao, Hai Wang, Pengwei Li, Wenkai Gao, Yu Zhang, Junli Chen, Nengqin Jia www.elsevier.com/locate/bios

PII: DOI: Reference:

S0956-5663(18)30679-1 https://doi.org/10.1016/j.bios.2018.08.070 BIOS10734

To appear in: Biosensors and Bioelectronic Received date: 12 June 2018 Revised date: 28 August 2018 Accepted date: 29 August 2018 Cite this article as: Pan Gao, Hai Wang, Pengwei Li, Wenkai Gao, Yu Zhang, Junli Chen and Nengqin Jia, In-site synthesis molecular imprinting Nb2O5 –based photoelectrochemical sensor for bisphenol A detection, Biosensors and Bioelectronic, https://doi.org/10.1016/j.bios.2018.08.070 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 galley proof before it is published in its final citable 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.

In-site synthesis molecular imprinting Nb2O5 – based photoelectrochemical sensor for bisphenol A detection Pan Gaoa, Hai Wangb, Pengwei Lia, Wenkai Gaoa, Yu Zhanga, Junli Chenb*, Nengqin Jiaa* a

The Education Ministry Key Laboratory of Resource Chemistry, Shanghai Key

Laboratory of Rare Earth Functional Materials, Department of Chemistry, Shanghai Normal University, Shanghai 200234, PR China b

School of Materials and Chemical Engineering, Collaborative Innovation Center of

Environmental Pollution Control and Ecological Restoration, Zhengzhou University of Light Industry, Zhengzhou 450001, PR China.

[email protected] [email protected]. *

Corresponding Authors: Tel/Fax: +86-21-64321833.

Abstract

In this work, a photoelectrochemical (PEC) sensor based on inorganic surface molecular imprinting Nb2O5 (MI-Nb2O5) for detection of bisphenol A (BPA) had been developed. In the PEC sensor, MI-Nb2O5 material was synthesized based on an in-situ surface molecular imprinting technique. The microstructure characteristics of

the as-prepared photoactive materials were systematically investigated by XRD, SEM, TEM, XPS, FTIR and UV-vis spectroscopy. The PEC detection results showed that the MI-Nb2O5 material had higher photocurrent responses and excellent selectivity for contaminant BPA under UV-light irradiation owing to the abundant special recognition sites on the surface of MI-Nb2O5. Besides, the PEC sensor exhibited a wide detection range from 0.01 nmol·L-1 to 30 nmol·L-1 with a low limit of detection (LOD) of 0.004 nmol·L-1. The interferences test showed that the sensor had a good selectivity to BPA molecules in the different interference solutions. This method combining molecular imprinting technique with photoelectrochemical detection measurement made a successful attempt to detect BPA and supplied a promising way to detect other environment pollutions rapidly and selectively in the future. Keywords Molecular imprinting technique, Nb2O5, photoelectrochemical detection

1. Introduction Bisphenol A (BPA, 2,2-bis (4-hydroxyphenyl) propane), an important organic chemical precursor, is widely used for the preparation of polycarbonate products, epoxy resins, polysulfone resin, polyphenylene oxide resin and so on. (Qiao et al.

2016; Yang et al. 2018c) Besides, BPA is also one of the endocrine disrupting compounds, which can affect the reproduction of aquatic organisms and cause a different of unfavorable health problems to human beings. (Deng et al. 2017; Han et al. 2015; Xu et al. 2017) Recently, BPA had been found widely in surroundings and food due to the leakage or hydrolysates from polycarbonate plastics, epoxy resins and other products. (Fu and Kawamura 2010; Oh et al. 2015) Owing to its severe threat to environment and human health, it is significant to develop a more convenient, higher sensitivity and specific method to detect BPA rapidly. In the last decades, many different analytical techniques have been applied to detect BPA, such as high performance liquid chromatography (HPLC), (Lin et al. 2011) gas chromatography-mass spectrometry (GC-MS), (Becerra and Odermatt 2012; Zhang et al. 2011) gas chromatography (GC), (Vandenberg et al. 2007) enzyme-linked immunosorbent assay (ELISA), (Feng et al. 2009) surface-enhanced Raman scattering (SERS), (Yang et al. 2018b) chemiluminescence immunoassay and so on. (Cao et al. 2014) However, these methods usually need expensive testing equipments, complex pre-treatment progress and qualified personnel, which make these high cost and time-consuming analytical methods be limited to detected BPA. Photoelectrochemical (PEC) sensing, a newly emerged and vibrantly developing analytical method based on photoinduced electron-hole transfer processes, has attracted more and more attentions in recent years since it interacts optical and electrochemical techniques to detect the concentration of analytical molecules. (Chen et al. 2018; Liu et al. 2017b; Zhang et al. 2017a) Its different forms of excitation (light) and detection (current) make a lower

background signals and higher sensitivity. (Zhao et al. 2014) However, the selectivity of the traditional PEC sensors was not satisfactory because the strong oxidative species (hydroxyl radicals produced during photocatalysis) are so powerful that most pollutions could be oxidized. Though mounts of identify elements had been introduced into the conventional PEC analysis process such as aptamers or antibodies, (Hatef et al. 2012; Zhang et al. 2013) they could not work well or keep stability in extreme environments, (Justino et al. 2015) and these sensors were usually designed for one-time using. (Zheng et al. 2011) These disadvantages had limited its practical application tremendously. Therefore, Molecular imprinting technique (MIT), named “artificial antibody”, owing to create mounts of special shapes, sizes and functional groups as molecular recognition sites for special detection analysis, has been used widely in molecular detection field to enhance the selectivity of system.(Yin et al. 2018b) However, it was usually use organic matters as functional monomers during the traditional molecular imprinting process, such as pyrrole (Py), dopamine (DA) and so on. (Lu et al. 2013) The polymer might be decomposed under the light irritation and reduce the mechanical strength or the stability of system. (Lu et al. 2012) So preparing photoactive materials with recognition sites by inorganic molecular imprinting technique might overcome above disadvantages and exhibit a better detection property. (Xin et al. 2017) Niobium pentoxide (Nb2O5), as an important n-type semiconductor material with a wide band gap (Eg≈3.4 eV), (Liu et al. 2017a; Rani et al. 2014), could be induced to generate electron and hole under photoexcitation. It had been widely applied in

capacitors, (Li et al. 2016) solar cells, (Fang et al. 2011) sensing, (Zhang et al. 2012) and photocatalyst because of its inherent merits such as nontoxicity, good chemical stability and easily synthesized. (Chen et al. 2017; Zhang et al. 2009; Zhao et al. 2012) Besides, by integrating MIT to create inorganic recognition sites on the surface of Nb2O5, which might exert enormous potential for Nb2O5 in PEC detection field. In this work, we proposed a novel strategy (scheme 1) by combining surface molecular imprinting technique with PEC analysis method to fabricate the PEC sensor and selectively detect BPA. Owing to introduce BPA molecules as templates during the process of producing photoactive Nb2O5, the molecular imprinting recognize sites for BPA could be easily created on the surface of Nb2O5 materials after further treatment. Attributed to interacting photoelectrochemical detect method with molecular imprinting technique, the obtained PEC sensor showed a wide response range, low detection limit and satisfactory selectivity under optimal conditions.

2. Experimental 2.1. Materials and reagents All chemical reagents were analytical grade and used without further purification. Nb foils (0.2 mm thickness, 99.9% purity) were purchased from Zhichengjinshu establishment. Bisphenol A was purchased from the Sigma-Aldrich Co. Hydrofluoric acid (HF) solution, Diphenolic acid, hydroguinone, bisphenol B, resorcinol, catechol, 2,5-Di-tert-butylhydroquinone, potassium hexacyanoferrate, potassium ferrocyanide

and absolute ethanol were purchased from Tianjin Fengchuan Chemical Reagent Technologies Co. Phosphate-buffered solutions (PBS, pH=6.81) was prepared from disodium hydrogen phosphate (NaHPO4·12H2O, 50 mM) and potassium phosphate monobasic (KH2PO4, 50 mM). All solutions were gained using double distilled water. 2.2. Apparatus A 300 W Xe lamp (Beijing Prefect Light, Microsolar 300) with a band-pass filter (λ = 365 ± 30 nm) was used as the irradiation source. Scanning electronic microscope (SEM) images were gotten by scanning electron microscope (JEOL, JSM-7001F). X-ray diffraction (XRD) patterns were gained by a X-ray diffractometer (Bruker, D8) using Cu Kα radiation (λ=1.54056 Å) and X-ray power of 40 kV/20 mA. X-ray photoelectron spectroscopy (XPS) paraments were obtained from a Thermo Scientific ESCAlab 250Xi photoelectron spectrometer equipped with a monochromatic Al Kα source (λ = 1486.7 eV). UV-vis diffuse reflection spectra were recorded using a UV-vis spectrometer (Hitachi, U-3900H) by applying BaSO4 as a reference. Fourier transforming infrared spectra (FT-IR) were recorded using Bruker 70V FT-IR spectrometer. Electrochemical impedance spectroscopy (EIS) and PEC measurements were carried out with the CHI 660E electrochemical workstation (Shanghai Chenhua Instrument Co. Ltd., China) using the standard three-electrode cell system. EIS was carried out in the 0.1 mol·L-1 KCl solution containing 2.5 mmol·L-1 [Fe(CN)6]3-/4- with a frequency from 0.1 Hz to 10.0 KHz at the open circuit potential. All PEC measurements were performed in phosphate-buffered solutions under ultraviolet light irradiation, and the basic voltage was set 0.3 V.

2.3. In-situ fabrication of MI-Nb2O5 PEC sensor Before preparation, Nb foils were sonicated in acetone, ethanol and distilled water for 10 min, respectively, followed by drying at 60 ℃. Then, 50 mL distilled water and 50 mL ethanol containing a certain amount of BPA molecules were stirred by a magnetic bar. Following, 100 μL HF solution was added into the above solution to adjust the pH of the mixed solution to 3.00 and further stirred at room temperature with 1 h. After that, the mixture and the prepared Nb foil were transferred into the Teflon-lined stainless steel autoclave and kept 200 ℃ for 12 h. In this hydrothermal process, the BPA template molecules could be adsorbed on the surface of Nb2O5. Then, the Nb foil covered with reaction products was washed by ethanol/ distilled water several times and dried at 60 ℃. Subsequently, these materials were further annealed at 500 ℃ for 2 h under N2 atmosphere to crystallize Nb2O5, meanwhile, the template molecules (BPA) could be removed from the surface of Nb2O5 by annealing treatment and left the special recognize sites. (Zhao et al. 2016) The final material was denoted as MI-Nb2O5. For comparison, non-imprinted Nb2O5 (denoted as NI-Nb2O5) was synthesized under the same conditions without adding BPA as template molecules during the hydrothermal process. 2.4. PEC performances of the MI-Nb2O5 sensor All photoelectrochemical measurements were carried out using the traditional three-electrode cell system on CHI 660E workstation. The MI-Nb2O5 or NI-Nb2O5 photoactive materials, Pt sheet and Ag/AgCl electrode were used as working electrode, counter electrode and reference electrode, respectively. The working electrode was

irradiated by a 300 W Xe lamp (Beijing Prefect Light, Microsolar 300) with a band-pass filter (λ = 365 ± 30 nm) horizontally. Amperometric current-time (I-t) measurement was adopted for both sensitivity and selectivity measurements at the bias potential of 0.3 V. The results of photocurrent responses at the different bias potential were shown in Fig S1.

3. Results and discussion 3.1. Characterizations of photoactive materials Scanning electron microscope (SEM) and transmission electron microscope (TEM) were carried out to investigate the morphological and structural properties of the photoactive materials. As SEM images of MI-Nb2O5 and NI-Nb2O5 (for comparison) showed in Fig. 1A and 1C, respectively. The MI-Nb2O5 and NI-Nb2O5 all exhibited nanospheres with uniform structure closely connected together and formed mesoporous structures, which is helpful to enlarge the adsorption capacity of BPA molecules and accelerate the diffusion and transportation of the target molecules. Meanwhile the diameters of the nanospheres were range from 70 to 120 nm with the average value about 90 nm. Interestingly, it was obviously seen that the MI-Nb2O5 nanospheres were composed of disorderly conglobatus Nb2O5 nanorods and the lengths of Nb2O5 nanorods were about 30 nm with an average diameter about 13 nm (Fig. 1B), the fringe spacing of 0.39 nm could be corresponded to the (001) crystal plane of Nb2O5 (shown in Fig. S2A), which indicated that nanorods grew along the [001] direction. (Li et al. 2016) However, the NI-Nb2O5 nanospheres (Fig. 1D and Fig.

S2B) were consisted of irregular nanoparticles rather than nanorods. Comparing with the fabrication process to form the two different morphology, the only difference was that adding BPA as templates would produce the MI-Nb2O5, while it would develop the NI-Nb2O5 without the addition of BPA under the same condition. It could be speculated that the discontinuous lattice fringes might arise from adsorbed BPA oxidation to extract surficial lattice oxygen, which could create lots of specific binding sites with efficient memory of the shapes, sizes, or functional groups of BPA molecules in the surface of Nb2O5 nanorods. Meanwhile, owing to without BPA adsorbed on the surface of materials when produced NI-Nb2O5, there were not nanorods formed and recognition sites could not be created on the surface of nanoparticles. Besides, X-ray diffraction (XRD) was applied to demonstrate the crystalline information about different Nb2O5 materials. The related spectra were shown in Fig.1E. The diffraction peaks at 38.60°, 55.75°, and 69.90° marked by asterisk corresponded to Nb foil substrate. The characteristic peaks of MI-Nb2O5 (without anneal) and NI-Nb2O5 (without anneal) located at 2θ = 22.52° and 46.14° were related to the diffractions from (001) and (002) planes of the hexagon phase of Nb2O5 (PDF# 28-0317),(Yan and Xue 2008) which indicated that BPA molecules appeared in the precursor solution during hydrothermal process could not change the crystal structure of Nb2O5 photoactive materials. After annealing treatment (curve d and e), the peaks of MI-Nb2O5 and NI-Nb2O5 located around 22.60°, 28.31°, 36.59°, 45.06° and 54.91° were indexed to (001), (180), (181), (002) and (182) planes of orthorhombic Nb2O5

(PDF# 27-1003).(Murayama et al. 2014) It was clearly observed that the intensity of (001) planes appeared in MI-Nb2O5 (without anneal) and MI-Nb2O5 were stronger than those in NI-Nb2O5 (without anneal) and NI-Nb2O5, which indicated that BPA molecules might act as organic directing agents and cause Nb2O5 material had a preferential orientation along the (001) plane. (Zhao et al. 2012) The results well matched with that exhibited by SEM and TEM results. Moreover, the intensities of the diffraction peaks were stronger and sharper after annealing, which indicated the nanomaterials

had a

better crystalline structure

and could

enhance the

photoelectrochemical responses. (Liu et al. 2017a) To evaluate the surface functional groups of different Nb2O5 materials during the process of fabricating PEC sensor, Fourier transforming infrared (FT-IR) spectra were recorded. As shown in Fig. 2A, BPA (curve a) and MI-Nb2O5 (without anneal) (curve b) showed characteristic peak at 1608 cm-1 and 1645 cm-1 (aromatic rings framework vibration), 2851 cm-1 and 2960 cm-1 (C-H vibration), (Giarola et al. 2017; Yin et al. 2018a) revealing that the BPA molecules were successfully grafted on the surface of Nb2O5. The peaks located at the 3432 cm-1 and 1645 cm-1 could be attributed to the stretching and bending mode of surfacial O-H groups, (Chai et al. 2012; Giarola et al. 2017; Rebocho et al. 2018) which were helpful to bind BPA molecules via hydrogen bonding and electrostatic interactions. The characteristic peaks were disappeared in the spectrum of MI-Nb2O5, indicating that BPA molecules were removed from the surface of Nb2O5 materials and abundant special active sites for the BPA might be left. Besides, in order to explore the PEC fabricating process, the photoelectrochemical

responses of the different electrode materials were recorded by amperometric method. The measurements were carried out in PBS solution at optimal conditions. As shown in Fig. 2B, bare Nb foil hardly had photocurrent response, while the photocurrent response values of NI-Nb2O5 and MI-Nb2O5 decreased distinctly to 1.01 μA·cm-1 and 0.95 μA·cm-1 from 1.19 μA·cm-1 and 1.51 μA·cm-1 when dipped into a certain of BPA solution. These photocurrent responses could evidence that the photoactive materials (MI-Nb2O5 or NI-Nb2O5) were synthesized successfully on the Nb foil. The differences of photocurrent density (△I=I0-I, where I0 and I were photocurrent density in PBS solution and 30 nmol·L-1 BPA solution, respectively) were showed in Fig. 2C. The △I value of MI-Nb2O5 (0.56 μA·cm-1) was nearly three times as large as that of NI-Nb2O5 (0.18 μA·cm-1). The results exhibited that the MI-Nb2O5 semiconductor materials could have a better photoelectrochemical response property and more sensitive to BPA molecules compared with NI-Nb2O5. It could be explained that after annealing treatment, there were some special recognition sites left on the surface of MI-Nb2O5, which could selectively adsorb BPA and cause the change of photocurrent obviously when met BPA molecules again. (Li et al. 2017; Nandy Chatterjee et al. 2017) It also could be speculated that the MI-Nb2O5 PEC sensor might be fabricated successfully. 3.2. Photoelectrochemical detection mechanism of MI-Nb2O5 PEC sensor The illustration of MI-Nb2O5 PEC sensor was showed in scheme 1. There were BPA template molecules adsorbed on the surface of Nb2O5 nanorods. after annealing treatment, BPA molecules were removed meanwhilevast of recognition sites were left.

Under ultraviolet light illumination, MI-Nb2O5 nanorods could be gave rise to the separation of electron and hole pairs, then the electrons transfer from valance band (VB) to conduction band (CB), which may generate photocurrents. However, when the BPA molecules were adsorbed to the surface, the organics might hinder the electron-transfer greatly and result in the recombination of electron and hole pairs, which ledto the noteworthy decrease in the photocurrent. (Hao et al. 2017; Wu et al. 2016) Based on the inhibition effect of BPA organic molecules on the PEC responses of MI-Nb2O5, the high sensitivity and selectivity sensor could be developed. X-ray photoelectron spectroscopy (XPS) was conducted to further exam the compositions and chemical states of the MI-Nb2O5 PEC sensor. The binding energy of 284.6 eV had been regarded as the standard of the chemical shift for C1s from the adventitious hydrocarbon. From the results showed in Fig. S3B, there were different characteristic peaks at 284.2 eV (C=C), 284.6 eV (C-C), 286.0 eV (C-O), 288.5 eV (O-C=O) of MI-Nb2O5 (without anneal), (An et al. 2014; Zhao et al. 2016) while the C=C characteristic peak disappeared after the annealing treatment to form MI-Nb2O5. The binding energies of MI-Nb2O5 (without anneal) were located at 209.6 eV and 206.9 eV corresponding to the 3d5/2 and 3d3/2 states of Nb5+ ions, respectively (Fig. S3C). (Kulkarni et al. 2017; Ma et al. 2018) Notably, for the MI-Nb2O5 photoactive material, except for the characteristic peaks at 209.8 eV and 207.0 eV matched well with the Nb-3d5/2 and Nb-3d3/2 of the Nb5+ ions, the peaks located at 209.2 eV and 206.3 eV were well corresponded to the Nb4+ ions. (Li et al. 2018) The percentage composition of Nb4+ in MI-Nb2O5 was about 26.4%, which indicated that the surficial

Nb5+ ions were partly reduced to Nb4+ by BPA molecules during the annealing process and abundant oxygen vacancy sites would be created. These imprinted specific sites would have a strong ability to selectively recognize BPA molecules from the multicomponent mixtures. From the Fig. S3D, the three different peaks at 529.9 eV, 531.0 eV, 532.1 eV were assigned to Nb-O bonds, -OH and C-O on the MI-Nb2O5 (without anneal), respectively. However, there were only two characteristic peaks at 529.7 eV (Nb-O bonds) and 531.1 eV(-OH) left after annealing treatment. (Ma et al. 2018) These results implied that BPA molecules may exist on the surface of MI-Nb2O5 (without anneal) and be removed successfully after annealing treatment. 3.3. Photocurrent responses of the MI-Nb2O5 PEC sensor The constructed MI-Nb2O5 PEC sensor was used to detect the diverse concentrations of BPA solutions. The obtained I-t curves (Fig. 3A) exhibited the change of photocurrent density was depended on different concentrations of BPA solutions and gradually decreased along with the increase of BPA concentrations. The change of photocurrent density was linearly proportional to the logarithm of the BPA concentration over the range of 0.01 nmol·L-1 to 30.0 nmol·L-1 (Fig. 3B). The linear regression equations were △I/I0 = 0.2804 + 0.0651 LogC with a correlation coefficient (R2) of 0.9943. Besides, the limit of detection (LOD) was estimated to be 0.004 nmol·L-1 (S/N=3). Table S1 had listed the linear range and LOD of different sensors in the reported literature. Compared with other assay methods, the prepared PEC sensor had a relatively wide linear range and lower LOD. The possible reasons for the improved selectivity in this MI-Nb2O5 PEC sensor were deduced: (1) BPA template

molecules were in-situ molecularly imprinted on the surface of Nb2O5 nanomaterial during the hydrothermal process, and abundant special recognition sites (special shapes, sizes and functional groups) for BPA molecules might be created on the surface of Nb2O5 nanorods after annealing treatment; (2) The special sites would increase the readsorption ability for BPA molecules when the BPA appeared again and lead to a wide concentration range for BPA detection. (Foguel et al. 2017; Justino et al. 2015) The PEC sensor showed a wider response range and a lower limit detection, suggesting the promising application by integrating PEC with molecular imprinting technology for fast detection of other contaminants of emerging concern. 3.4. Selectivity and stability of the developed PEC sensor To investigate the selectivity of the MI-Nb2O5 PEC sensor, several other chemicals, including diphenolic acid, hydroguinone, bisphenol B, resorcinol, pyrocatechol and 2,5-Di-tert-butylhydroquinone (denoted as S1-S6) were chosen as interferences. The change of photocurrent density after adding these interfering substrates was recorded using the amperometric technique. The selectivity had been studied by recording the photocurrent density in different interfering solutions. The R (%) value was evaluated by the following equation: R(%) =

𝐼𝐵𝑃𝐴 -𝐼 × 100% (1) 𝐼𝐵𝑃𝐴

where IBPA was the photocurrent density in 30 nmol·L-1 BPA solution, and I was the photocurrent density in 30 nmol·L-1 BPA containing 3 mmol·L-1 interfering substrates solution (100-fold concentration of BPA), respectively. The R (%) value for 30 nmol·L-1 BPA was set to be 100%, and the R (%) values for other mixture solution

were calculated by Eq. (1). For comparing, NI-Nb2O5 materials were tested under the same condition. The insert of Fig. 4A showed clearly that the photocurrent of MI-Nb2O5 was lower than that of NI-Nb2O5 in different solutions, which indicated that BPA molecules were adsorbed to the MI-Nb2O5 successfully and caused the photocurrent to decrease. The R (%) values of MI-Nb2O5 were less than 20% in the interference solutions and some of these values even less than 10%, which could be accepted. However, the organic matters (BPA and other interferences) could not be adsorbed to the materials by hydrogen bonding or electrostatic interactions owing to the fact that there were not abundant functional groups on the surface of NI-Nb2O5, which could not cause the photocurrent decreased in different solutions. Therefore, the MI-Nb2O5 PEC sensor had an excellent selectivity for BPA molecules due to the specificity of molecular imprinting recognition sites. The stability was an important parameter to evaluate the MI-Nb2O5 PEC sensor during the detection process. In this work, the stability of MI-Nb2O5 PEC sensor was examined by recording the photocurrent continuously in the solution. As shown in Fig. 4B, the photocurrent responses remained relatively stable with a decrease about 5% after several on/off irradiation cycles for 800 s. And the photocurrent intensity only decreased less than 7% after three months stored at room temperature, indicating that the fabricated MI-Nb2O5 PEC sensor had a long-term stability and good repeatability for BPA molecules detection. 3.5. Preliminary analysis of real sample by MI-Nb2O5 PEC sensor In order to evaluate the feasibility of the fabricated PEC sensor in practical

application, the real sample of water was purchased from a local supermarket. The recovery experiments of BPA were carried out by the standard addition methods in real water sample and the results were presented in Table 1. The recovery (between 89 and 120%) and the accuracy by calculating the RSD (between 1.7 and 5.4%) were acceptable, which demonstrated that the prepared MI-Nb2O5 PEC sensor had an excellent accuracy and could be used for detecting the concentration of BPA solution in real water sample.

4. Conclusions In this work, a photoelectrochemical sensor based on an in-situ inorganic surface molecular imprinting Nb2O5 (MI-Nb2O5) for detection of bisphenol A had been prepared. Detailed characteristic tests showed that the MI-Nb2O5 materials had a nanorod structure; BPA molecules could be grafted and removed successfully from nanomaterials, which could create mounts of recognition sites on the surface of MI-Nb2O5. PEC detection results exhibited the fabricated sensor had a relatively wide detection range from 0.01 nmol·L-1 to 30 nmol·L-1 with a low limit of detection (0.004 nmol·L-1). Besides, interferences tests showed that the PEC sensor had a good selectivity for BPA molecules in different interference solutions. This work using molecular imprinting technique with photoelectrochemical detection measurement made a successful attempt to detect BPA, which supplied a promising way to detect other environment pollutions rapidly and selectively. There are still need more work

for further research in the future.

Conflicts of interest There are no conflicts to declare.

Acknowledgements The authors are grateful to the financial support from the Shanghai Science and Technology Committee (17070503000), National Natural Science Foundation of China (21373138) Program for Changjiang Scholars and Innovative Research Team in University (IRT_16R49) and International Joint Laboratory on Resource Chemistry (IJLRC). Key Program of Henan Province for Science and Technology (16A150047), School

Doctor

Foundation

of

Zhengzhou

University

of

Light

Industry

(2012BSJJ014).

References An, X., Li, K., Tang, J., 2014. ChemSusChem 7(4), 1086-1093. Becerra, V., Odermatt, J., 2012. Analyst 137(9), 2250-2259. Cao, W., Chao, Y., Liu, L., Liu, Q., Pei, M., 2014. Sens. Actuators B-Chem. 204, 704-709. Chai, S., Zhao, G., Zhang, Y.N., Wang, Y., Nong, F., Li, M., Li, D., 2012. Environ. Sci. Technol. 46(18), 10182-10190. Chen, J., Gao, P., Wang, H., Han, L., Zhang, Y., Wang, P., Jia, N., 2018. J. Mater.

Chem. C 6(15), 3937-3944. Chen, J., Wang, H., Huang, G., Zhang, Z., Han, L., Song, W., Li, M., Zhang, Y., 2017. J. Alloys Compd. 728, 19-28. Deng, K., Liu, X., Li, C., Hou, Z., Huang, H., 2017. Anal. Methods 9(37), 5509-5517. Fang, X., Hu, L., Huo, K., Gao, B., Zhao, L., Liao, M., Chu, P.K., Bando, Y., Golberg, D., 2011. Adv. Funct. Mater. 21(20), 3907-3915. Feng, Y., Zhou, Y., Zou, Q., Wang, J., Chen, F., Gao, Z., 2009. J. Immunol. Methods 340(2), 138-143. Foguel, M.V., Pedro, N.T.B., Wong, A., Khan, S., Zanoni, M.V.B., Sotomayor, M., 2017. Talanta 170, 244-251. Fu, P., Kawamura, K., 2010. Environ. Pollut. 158(10), 3138-3143. Giarola, J.d.F., Borges, K.B., Tarley, C.R.T., de Oliveira, F.M., Ribeiro, E.S., Pereira, A.C., 2017. Arab. J. Chem. 10(3), 430-438. Han, J., Qiu, W., Tiwari, S., Bhargava, R., Gao, W., Xing, B., 2015. J. Mater. Chem. A 3(16), 8870-8881. Hao, N., Chen, S., Qian, J., Zhang, Y., Liu, Q., Zhang, X., Wang, K., 2017. J. Electroanal. Chem. 801, 410-415. Hatef, A., Zare, A., Alavi, S.M., Habibi, H.R., Linhart, O., 2012. Environ. Toxicol. Chem. 31(9), 2069-2077. Justino, C.I.L., Freitas, A.C., Pereira, R., Duarte, A.C., Rocha Santos, T.A.P., 2015. TrAC, Trends Anal. Chem. 68, 2-17. Kulkarni, A.K., Praveen, C.S., Sethi, Y.A., Panmand, R.P., Arbuj, S.S., Naik, S.D.,

Ghule, A.V., Kale, B.B., 2017. Dalton Trans. 46(43), 14859-14868. Li, D., He, Q., He, Y., Xin, M., Zhang, Y., Shen, Z., 2017. Biosens. Bioelectron. 94, 328-334. Li, H., Zhu, Y., Dong, S., Shen, L., Chen, Z., Zhang, X., Yu, G., 2016. Chem. Mater. 28(16), 5753-5760. Li, R., Qin, Y., Liu, X., Yang, L., Lin, C., Xia, R., Lin, S., Chen, Y., Li, J., 2018. Electrochim. Acta 266, 202-211. Lin, C.Y., Fuh, M.R., Huang, S.D., 2011. J. Sep. Sci. 34(4), 428-435. Liu, M., Ding, X., Yang, Q., Wang, Y., Zhao, G., Yang, N., 2017a. J. Hazard. Mater. 331, 309-320. Liu, X., Huo, X., Liu, P., Tang, Y., Xu, J., Liu, X., Zhou, Y., 2017b. Electrochim. Acta 242, 327-336. Lu, B., Liu, M., Shi, H., Huang, X., Zhao, G., 2013. Electroanalysis 25(3), 771-779. Lu, C.H., Zhang, Y., Tang, S.F., Fang, Z.B., Yang, H.H., Chen, X., Chen, G.N., 2012. Biosens. Bioelectron. 31(1), 439-444. Ma, X., Chen, Y., Lee, J., Yang, C., Cui, X., 2018. New J. Chem. 42(2), 1300-1308. Murayama, T., Chen, J., Hirata, J., Matsumoto, K., Ueda, W., 2014. Catal. Sci. Technol. 4(12), 4250-4257. Nandy Chatterjee, T., Banerjee Roy, R., Tudu, B., Pramanik, P., Deka, H., Tamuly, P., Bandyopadhyay, R., 2017. Sens. Actuators B-Chem. 246, 840-847. Oh, W.-D., Dong, Z., Hu, Z.-T., Lim, T.-T., 2015. J. Mater. Chem. A 3(44), 22208-22217.

Qiao, Y., Li, J., Li, H., Fang, H., Fan, D., Wang, W., 2016. Biosens. Bioelectron. 86, 315-320. Rani, R.A., Zoolfakar, A.S., O'Mullane, A.P., Austin, M.W., Kalantar-Zadeh, K., 2014. J. Mater. Chem. A 2(38), 15683-15703. Rebocho, S., Cordas, C.M., Viveiros, R., Casimiro, T., 2018. The J. Supercrit. Fluid. 135, 98-104. Vandenberg, L.N., Hauser, R., Marcus, M., Olea, N., Welshons, W.V., 2007. Reprod. Toxicol. 24(2), 139-177. Wu, W., Xiao, H., Luo, S., Liu, C., Tang, Y., Yang, L., 2016. Sens. Actuators B-Chem. 222, 747-754. Xin, Y., Li, Z., Wu, W., Fu, B., Wu, H., Zhang, Z., 2017. Biosens. Bioelectron. 87, 396-403. Xu, L., Yan, P., Li, H., Ling, S., Xia, J., Xu, Q., Qiu, J., Li, H., 2017. RSC Adv. 7(13), 7929-7935. Yan, C., Xue, D., 2008. Adv. Mater. 20(5), 1055-1058. Yang, F., Zhu, X., Fang, J., Chen, D., Feng, W., Fang, Z., 2018a. Ceram. Int. 44(6), 6918-6925. Yang, L., Chen, Y., Shen, Y., Yang, M., Li, X., Han, X., Jiang, X., Zhao, B., 2018b. Talanta 179, 37-42. Yang, Q., Wu, X., Peng, H., Fu, L., Song, X., Li, J., Xiong, H., Chen, L., 2018c. Talanta 176, 595-603. Yin, W., Wu, L., Ding, F., Li, Q., Wang, P., Li, J., Lu, Z., Han, H., 2018a. Sens.

Actuators B-Chem. 258, 566-573. Yin, Z.Z., Cheng, S.W., Xu, L.B., Liu, H.Y., Huang, K., Li, L., Zhai, Y.Y., Zeng, Y.B., Liu, H.Q., Shao, Y., Zhang, Z.L., Lu, Y.X., 2018b. Biosens. Bioelectron. 100, 565-570. Zhang, F., Zhang, P., Wu, Q., Xiong, W., Kang, Q., Shen, D., 2017a. Electrochim. Acta 247, 80-88. Zhang, H., Wang, Y., Yang, D., Li, Y., Liu, H., Liu, P., Wood, B.J., Zhao, H., 2012. Adv. Mater. 24(12), 1598-1603. Zhang, J., Cooke, G.M., Curran, I.H., Goodyer, C.G., Cao, X.L., 2011. J. Chromatogr. B 879(2), 209-214. Zhang, Q., Ge, J., Pham, T., Goebl, J., Hu, Y., Lu, Z., Yin, Y., 2009. Angew. Chem. Int. Ed. Engl. 48(19), 3516-3519. Zhang, Y., Cao, T., Huang, X., Liu, M., Shi, H., Zhao, G., 2013. Electroanalysis 25(7), 1787-1795. Zhao, W., Zhao, W., Zhu, G., Lin, T., Xu, F., Huang, F., 2016. Dalton Trans 45(9), 3888-3894. Zhao, W.W., Xu, J.J., Chen, H.Y., 2014. Chem. Rev. 114(15), 7421-7441. Zhao, Y., Eley, C., Hu, J., Foord, J.S., Ye, L., He, H., Tsang, S.C., 2012. Angew. Chem. Int. Ed. 51(16), 3846-3849. Zheng, Z., Zhou, Y., Li, X., Liu, S., Tang, Z., 2011. Biosens. Bioelectron. 26(6), 3081-3085.

Scheme 1. The illustration of PEC detection mechanism.

Figure 1. SEM image (A) and TEM image (B) of MI-Nb2O5, SEM image (C) and TEM images (D) of NI-Nb2O5. (E) XRD patterns of different materials (from a to e:

bare Nb foil, NI-Nb2O5 (without anneal), MI-Nb2O5 (without anneal), NI-Nb2O5, MI-Nb2O5).

Figure 2. (A) FT-IR spectra of BPA (a), MI-Nb2O5 (without anneal) (b), MI-Nb2O5 (c). (B) Photocurrent responses of Nb foil (a), NI-Nb2O5 (b), MI-Nb2O5 (d) in PBS solution and NI-Nb2O5 (c), MI-Nb2O5 (e) in PBS solution containing 30 nmol·L-1 BPA at the basis of 0.3 V. (C) the difference of photocurrent density of NI-Nb2O5 and MI-Nb2O5.

Figure 3. (A) Photocurrent responses of PEC sensor in BPA solution with different concentrations: (from a to i) 0, 0.01 nmol·L-1, 0.03 nmol·L-1, 0.05 nmol·L-1, 0.1 nmol·L-1, 1.0 nmol·L-1, 3.0 nmol·L-1, 10.0 nmol·L-1, 30.0 nmol·L-1. (B) The linear relationship between △I/I0 and the logarithm of BPA concentration from 0.01 nmol·L-1 to 30.0 nmol·L-1.

Figure 4. (A) Relative photocurrent density ratio of MI-Nb2O5 (Black) and NI-Nb2O5 (Red) in different interfering solutions (The insert was the photocurrent density of MI-Nb2O5 and NI-Nb2O5 in mixture solutions, S1 to S6: diphenolic acid, hydroguinone,

bisphenol

B,

resorcinol,

pyrocatechol

and

2,5-Di-tert-butylhydroquinone). (B) The stability of the MI-Nb2O5 PEC sensor in PBS solution.

Table 1. Detection of BPA in real water sample using MI-Nb2O5 PEC sensor. Added

Founded

Recovery

RSD

(nmol·L-1)

(nmol·L-1)

(%)

(%, n=3)

1

0

0.005

2

0.01

0.012

120

3.9

3

0.1

0.089

89

5.4

4

1

1.131

113.1

2.2

5

10

10.748

107.5

1.7

Samples

4.8

Highlights: 1. Used molecular imprinting technique to enhance the selectivity of PEC system. 2. Inorganic MIT overcame the disadvantages in traditional PEC detection filed. 3. In-site synthesis method simplified the complex pretreatment process greatly. 4. This method supplied a promising way to detect other pollutions in the future.