Tao Liu, Ying Xie, Lei Shi, Yu Liu, Zhenyu Chu*, Wanqin Jin*

State Key Laboratory of Materials-Oriented Chemical Engineering, College of Chemical Engineering, Nanjing Tech University, Nanjing 211816, China

Keywords:Prussian blue Carbon nanofiber Pt nanoparticle 3D architecture High sensitivity

ABSTRACT Nowadays, water pollution has become more serious, greatly affecting human life and healthy.Electrochemical biosensor, a novel and rapid detection technique, plays an important role in the realtime and trace detection of water pollutants.However, the stability and sensitivity of electrochemical biosensors remain a great challenge for practical detections in real samples to the strong interferences derived from complex components and coagulation effects.In this work, we reported a novel threedimensional architecture of Prussian blue nanoparticles (PBNPs)/ Pt nanoparticles (PtNPs) composite film, using 3D interweaved carbon nanofibers as a supporting matrix, for the construction of screenprinted microchips-based biosensor.PtNPs with diameters of ～2.5 nm was highly dispersed on the carbon nanofibers(CNFs)to build a 3D skeleton nanostructure through a solvothermal reduction.Subsequently,uniform PBNPs were in-situ self-assembled on this skeleton to construct a 3D architecture of PB/Pt-CNF composite film.Due to the synergistic effects derived from this special feature, the as-prepared hydroquinone (HQ) biosensor chips can synchronously promote both surface area and conductivity to greatly enhance the electrocatalysis from enzymatic reaction.This biosensor has exhibited a high sensitivity of 220.28 μA·L·mmol-1·cm-2 with an ultrawide linear range from 2.5 μmol·L-1 to 1.45 mmol·L-1 at a low potential of 0.15 V, as well as the satisfactory reproducibility and usage stability.Besides, its accuracy was also verified in the assays of real water samples.It is highly expected that the 3D PB/Pt-CNF based screen-printed microchips will have wide applications in dynamic monitoring and early warning of analytes in the various practical fields.

1.Introduction

Phenolic compounds, a kind of hazardous, non-degradable and carcinogenic chemical, have brought serious injury to the environment and human bodies [1].Generally, phenolic substances pollution is mainly originated from industrial wastewater discharge in the past few years due to the rapid productivity development [2,3].According to the integrated wastewater discharge standard of China, the maximum permitted concentration of phenolic substances is 0.4 mg·L-1[4].Hence, the rapid and on-time monitoring of phenolic compounds in water is critical to prevent excess waste discharge and pollutant leakage.At present, the most commonly used detection techniques are high-performance liquid chromatography (HPLC)[5], gas chromatography/mass spectrometry (GC/MS) [6], etc.However, most of them are unable to realize the real-time and in-situ analysis of phenolic compounds owing to the complex pre-treatment process and long detection period [7].

In recent years,electrochemical biosensors have received enormous interests due to their rapid response,on-line detection mode and high specificity inherent in enzymatic reactions, which transduce the desired information of target analytes into electrochemical signals [8].As to phenolic compound detection, laccase is an effective biocatalyst for the oxidation reaction,producing electrons transferring to the electrode [9].However, various phenolic compounds often coexist in real water system and will be oxidated at their specific oxidation potential(e.g.hydroquinone:～0.15 V,catechol:～0.25 V)[10,11].Therefore,it is essential and difficult to distinguish the response signals from different phenolic compounds,which mainly depends on the electrocatalysis activity of the electrode.In addition, the stability of phenolic biosensors is limited due to the easy inactivation of laccase and serious aggregation of sensing materials, causing poor reproducibility and unsatisfactory activity of enzymatic reaction on the electrode interface.

To solve the above problems, novel nanomaterials and nanostructures are applied to optimize enzyme loading and reduce the risks of losing enzymatic activity.Prussian blue (PB) has been served as an electron transfer mediator due to its excellent electrocatalytic reduction even at a low potential[12].As a result,PB and PB analogues based enzymatic biosensors have been widely applied in the assay of many biological analytes, e.g.glucose, lactate and phenolics [13-16].Despite this impressive merit, there remains two highly concerned issues in PB based biosensors to obtain sensitive responses.On one hand, as a semiconductor with a wide band gap of ～1.4 eV, PB is of a poor electrical conductivity which creates a high electron transfer resistance to limit the signal transmission [17].On the other hand, the easy aggregation of PB nanocrystals often causes the decrease of active sites on their surfaces,which obviously weakens the ability of the signal generation.Metallic nanoparticles,especially gold,platinum and silver,usually possess ultrahigh conductivity, surface area and electrocatalysis[18-22], which are promising to cover the low conductivity of PB through compositing.Besides, to improve the dispersion of nanomaterials, a possible solution is to anchor these nanomaterials on a specific supporting matrix which is often composed by carbon materials, including graphene, carbon nanotubes (CNTs), etc.Carbon nanofibers (CNFs) are extremely attractive in the field of bioanalytical science, due to the superior specific area, electrical conductivity, biocompatibility and low cost [23,24].Importantly,CNFs have a much larger functionalized surface area compared to that of CNTs, and possess an improved dispersibility than that of graphene.Therefore,if CNFs can be served to provide a supporting matrix for PB growth accompanying with the assembly of noble metal nanoparticles, both conductivity and electrocatalysis of the prepared biosensor will be expected to improve simultaneously.

In this work,we designed a novel 3D architecture of PB/Pt-CNF based screen-printed microchip for the construction of an advanced hydroquinone biosensor.As shown in Fig.1, the monodispersed PtNPs were uniformly distributed on the acidification CNFs skeleton through a solvothermal reduction.The selfformed 3D skeleton of Pt-CNF composites was served as the matrix substrate to in-situ self-assemble PB nanoparticles(PBNPs)for promoting the electron transfer during the electrochemical reaction.After the laccase immobilization,the constructed biosensor exhibited a wide detection range under a low working potential with a high accuracy in the assay of hydroquinone even in real water samples, which are attributed to synergistic effects on conductivity,electrocatalysis and excellent stability of the 3D architecture for loading nanomaterials and enzymes.

2.Experimental

2.1.Materials and reagents

Hexachloroplatinic acid (H2PtCl6·6H2O), K4[Fe(CN)6]·3H2O,FeCl3·6H2O, and laccase (EC 1.10.3.2, 285 unit per mg, from Rhus vernicifera) were purchased from Sigma-Aldrich.Carbon nanofibers (CNFs) were supplied by Nanjing XFNANO Materials Tech Co.,Ltd.Glutaraldehyde 25%(volume)was obtained from Shanghai Lingfeng Chemical Reagent Co., Ltd.(China).Hydrogen peroxide(H2O2, 30%, solution), phenol (PE), resorcine (RC), hydroquinone(HQ) and catechol (CC) were received from Sinopharm Chemical Reagent Co., Ltd.(China).Other chemicals employed were all of analytical grade.The screen-printed electrodes (SPE) were selfprepared with carbon ink and plythylene terephthalate (PET) substrate (working electrode diameter: 3 mm).All aqueous solutions were prepared with deionized water (≥18.2 MΩ, Smart2Pure 6,Thermo Fisher Scientific, USA).

2.2.Fabrication of PtNPs decorated carbon nanofibers (Pt-CNFs)

Fig.1.Schematic illustration of the preparation of PB/Pt-CNF based SPE.The preparation process includes three successive steps: (A) the formation of 3D Pt-CNF skeleton through CNF acidification and in-situ synthesis of PtNPs; (B) Self-assembly of PBNPs on 3D skeleton and (C) Laccase loading on the PB/Pt-CNF for sensitive HQ detection.

The Pt-CNFs were prepared according to the reported literatures[25,26]with some modifications.In a brief, CNFs were chemically shortened in a concentrated H2SO4/HNO3acid solution and the residual impurities were removed at the same time (Fig.1).Then,25 mg H2PtCl6·6H2O was dissolved in a mixture solution of 60 ml EG and 40 ml water.Meanwhile,40 mg pretreated CNFs were dispersed in a mixture solution of 30 ml EG and 20 ml water,following by the slow addition of H2PtCl6·6H2O solution with continuous stirring.Subsequently, the obtained solution was sonicated for 15 min and then refluxed at 135 °C for 8 h to ensure complete reduction of Pt.After the reaction,the mixture solution was filtered and thoroughly washing with acetone,isopropyl alcohol and distilled water.Finally, the obtained Pt-CNFs were collected and dried under vacuum at 70 °C for 24 h.

2.3.Preparation and pre-treatment of screen-printed electrode

Polyethylene terephthalate (PET) plates were used as the substrate to support the three-electrode system via screen-printing.The working electrode (WE) and the counter electrode (CE) were printed by carbon ink while the reference electrode(RE)was fabricated by silver chloride ink.After the printing process,the PET substrate was dried and divided into several chips, of which the size was about 4(L)×1(W)cm.Then the SPE was rinsed by deionized water and dried for further use.

2.4.Preparation of PBNPs on the Pt-CNF skeleton modified electrode and SPE

Firstly,Au disk electrode(Diameter=2 mm)was polished with 0.3 and 0.05 μm alumina slurry on microcloth pads, successively sonicated with pure water and ethanol, and finally dried with nitrogen.Then, 5 μl of 0.2 mg·ml-1Pt-CNFs was dropped onto the electrode surface and dried with nitrogen to form a 3D skeleton structure.To prepare PB on the 3D skeleton, the self-assembly approach established in our previous researches is used due to its advantages of precise regulation on the PB morphology.Two precursor solutions consisting of 0.01 mol·L-1K4[Fe(CN)6],0.1 mol·L-1KCl, 0.1 mol·L-1HCl (solution A) and 0.01 mol·L-1FeCl3, 0.1 mol·L-1KCl, 0.1 mol·L-1HCl (solution B) were prepared respectively for the self-assembly preparation(Fig.1).During each assembly cycle, the Pt-CNF modified electrode was consecutively dipped into solution A for 60 s, then distilled water 3 times and 10 s each for cleaning,solution B for 60 s,and again distilled water 3 times for cleaning.Finally, the PB/Pt-CNF composite film on the electrode was obtained and dried with nitrogen.The preparation process of PBNPs on SPE was similar to the above procedures.

2.5.Construction of the HQ biosensor on SPE

Before the immobilization of laccase on PB/Pt-CNF electrode,100 U of laccase was dissolved in 0.5 ml phosphate buffer solution(PBS,0.1 mol·L-1, pH= 7.0),and 2 μl glutaraldehyde was added to crosslink the enzyme.Then,10 μl of the above laccase solution was dropped on the PB/Pt-CNF electrode and stored at 4 °C overnight,following by washing with distilled water and drying in nitrogen.

2.6.Characterizations and electrochemical measurements

Microstructure characterizations were implemented with a transmission electron microscope(TEM,JEOL2100),and field emission scanning electron microscope (FESEM, Hitachi, S-4800).The X-ray diffraction (XRD) of the above materials were performed on an X-ray diffractometer (D/MAX 2500 V/PC) with Cu Kα radiation (0.15419 nm).An X-ray photoelectron spectrometer (XPS)was implemented by ESCALAB 250 spectrometer (Thermo Scientific, Al Kα X-ray of 1486.6 eV as the light source).All the electrochemical measurements were carried out on a CHI 660E electrochemical workstation (Shanghai Chenhua Instrument Co.,Ltd., China).The three-electrode system used consists a PB based working electrode, a platinum wire counter electrode, a saturated silver-silver chloride electrode (Ag/AgCl) reference electrode.As to the SPE, other electrodes are not required to participate in the electrochemical processes.Cyclic voltammetry (CV) tests and amperometric detection were performed in 0.05 mol·L-1PBS(pH = 6.5) containing 0.1 mol·L-1KCl at room temperature.The scan rate in CV was 50 mV·s-1,and the applied potential in amperometric detection was -0.05 V.Electrochemical impedance spectroscopy (EIS) measurements were tested in 5 mmol·L-1K3[Fe(CN)6]/K4[Fe(CN)6]solution containing 0.1 mol·L-1KCl with the frequency changed from 0.1 Hz to 1000 kHz with a signal amplitude of 5 mV.

3.Results and Discussion

3.1.Structural evolution of Pt-CNFs and 3D PB/Pt-CNF film

The CNFs with outer/inside diameters of ～70/20 nm were used for anchoring PtNPs (Fig.2a).As shown in Fig.2b, PtNPs were homogenously loaded on the surface of CNFs, indicating the successful preparation of Pt-CNFs.Using a high magnification on the surface of one CNF,it can be found that the PtNPs with an average diameter of ～2.5 nm were well dispersed with rare aggregation.After dropping the Pt-CNFs on a bare electrode, the fibers were interlaced together to self-form a favorable 3D skeleton structure(Fig.2c), providing a significant increase of the surface area to act as a novel supporting matrix for the subsequent PB growth.By using a self-assembly approach, a mass of PBNPs has in-situ crystallized on the 3D skeleton with obvious porous structures produced by the stack of fibers (Fig.2d).The EDX results(Fig.2h-l) showed an obvious 3D skeleton structure with welldistributed PB and PtNPs.This obtained structure of PB/Pt-CNF film can be expected to not only improve its electrocatalytic activity and electrical conductivity,but also allows the target diffusion into the pores that provides abundant binding sites for the immobilization of enzymes.

XRD patterns in Fig.3a shows that the diffraction peak at 2θ=26.2° is attributed to the carbon nanofibers (0 0 2) crystal face of graphite (JCPDS 41-1487).After the deposition of PtNPs, the characteristic peaks located at ca.39.82° and 46.23° represents the (1 1 1) and (2 0 0) faces of Pt, respectively [8], which are in good agreement with the face-centered cubic(fcc)crystalline platinum (JCPDS 04-0802).In the first 10 cycles of PB self-assembly,the (2 0 0) plane of PBNPs was not observed immediately because of the low loading and irregular crystal of PB(Fig.2d).After 20 selfassembly cycles, more PBNPs were obtained on the CNF surface(Fig.2e), and until 30 cycles, obvious cube-like PB uniformly deposited on the 3D skeleton structure (Fig.2f-g).However, if 10 cycles were further assembled, more fibers showed connection together caused by the heavier aggregation of PB crystals.Meanwhile, almost PB crystals exhibited the irregular nanostructure decreasing the specific surface area of the film.Two peaks located at 17.69° and 24.86° are assigned to the (2 0 0) and (2 2 0) reflections of the PBNCs, respectively (JCPDS 73-0687).

In order to investigate the chemical composition and valence states,the nanocomposites were analyzed by XPS before and after Pt and PB deposition.Observed in the full survey spectra of Fig.3b,the PB/Pt-CNF contains the characteristic peaks that includes Pt,C,N, In (ITO substrate), O and Fe elements.Fig.3c shows that two main 4f peaks belonging to Pt locate at 70.98 eV (4f7/2) and 74.29 eV (4f5/2), representing the zero valent state of Pt.However,two satellite peaks with higher binding energy appear at 72.47 eV and 76.31 eV,which may be contributed by some few oxidized Pt,such as Pt(OH)2and PtO/PtO2[27-29].The Fe 2p spectra in Fig.3d reveals the characteristic peaks of Fe2+and Fe3+at 707.97 eV (Fe2+2p3/2),712.28 eV(Fe3+2p3/2),720.88 eV(Fe2+2p1/2)and 724.64 eV(Fe3+2p1/2), indicating the crystallization of PBNPs on the 3D framework [16].The main core-level peak of N1s (Fig.3e) can be fitted to three peaks at 397.2 eV, 399.2 eV and 402.1 eV, corresponding to the C-N group from [Fe(CN)6]4-in PB and PB/Pt-CNF nanocomposite [30,31].There were 6 peaks at 284.8 eV,285.2 eV, 285.7 eV, 286.7 eV, 287.3 eV and 289.0 eV in the C1s spectrum of PB/Pt-CNF(Fig.3f),which were attributed to graphitic carbon, C-C, C-N, C-O-C/C-O-H, C=O and O=C-O peaks,respectively[26,32].The existence of C-O-H and O=C-O enables to crosslink with enzyme, forming O-C=O and O=C-NH bonds,therefore greatly does benefits to the immobilization of GOx in the following experiment.

Fig.2.TEM images of(a)carbon nanofibers,(b)highly dispersed PtNPs on the CNFs;FESEM images of(c)3D Pt-CNF skeleton on ITO,(d-g)3D PB/Pt-CNF film prepared with different assembly cycles: (d) 10 cycles, (e) 20 cycles, (f) 30 cycles, (g) 40 cycles.; EDX mapping of (h and i) element overlay, (j-l) Fe, C and Pt element.

Fig.3.(a)XRD patterns of ITO substrate,CNF,PB,Pt-CNF and PB/Pt-CNF with different cycles;(b)XPS full survey spectra of Pt-CNF,PB and PB/Pt-CNF modified ITO;(c-f)XPS spectra of Pt4f, Fe2p, N1s and C1s.

CV tests were performed to investigate electrochemical features of above different modified electrodes (Fig.4a).Compared to the bare electrode,the introduction of CNF and PtNPs reduced the peak potential difference (ΔEp), causing a decrease in electron transfer resistance.However, only PB modified electrode showed a couple of higher redox peaks with a larger peak potential difference,indicating a significant improvement in electrocatalytic performance but a poor electron transfer ability.After introducing Pt-CNF to support the self-assembly of PB, the highest values of redox currents were observed, revealing that the composite of Pt-CNF and PB greatly enhanced the electrocatalytic activity of the electrode and effectively minimized the electron transfer resistance.

Fig.4.(a)CV behaviors of the bare,CNF,Pt-CNF,30 cycles of PB and 30-PB/Pt-CNF modified Au electrodes in 10 mmol·L-1[Fe(CN)6]3-/4-solution containing 0.1 mol·L-1 KCl with the scanning rate of 50 mV·s-1;(b)Nyquist plots of different modified electrodes in 5 mmol·L-1[Fe(CN)6]3-/4-solution containing 0.1 mol·L-1 KCl;(c)Calibration curves for the peak currents vs.the square root of scan rates(from 50 to 300 mV·s-1)with different modified electrodes in 10 mmol·L-1 K3Fe(CN)6 containing 3 mol·L-1 KCl;(d)CV behaviors of PB/Pt-CNF films assembled with various cycles in PBS buffer containing 0.1 mmol·L-1 H2O2;(e)Calibration curves for the peak currents vs.the square root of scan rates(from 50 to 300 mV·s-1)with different cycles of PB/Pt-CNF in 10 mmol·L-1 K3Fe(CN)6 containing 3 mol·L-1 KCl;(f)CV curves of PB/Pt-CNF in PBS with H2O2 for 30 cycles.

Meanwhile, EIS tests were adopted to investigate their interfacial electron transfer resistance (Rct) (Fig.4b).In comparsion with a Rctof ～150 Ω from the bare electrode, an obviously enlarged Retof ～165 Ω was observed to the only PB film, which was caused by the poor conductivity of PB.After the 3D achitecture of Pt-CNF film was adopted for PB deposition with 30 cycles,the Rctwas then significantly reduced to ～135 Ω which was derived from the high conductivities of both Pt and CNF as well as the large specific area for target diffusion and electron transfer.

In order to further confirm advantages of the 3D structure, the effective specific surface area of the modified electrodes were studied through the following Randles-Sevcik equation [4,33]:

In this equation, parameters of n, D0and C0are all constant values.

Hence, there is a linear relationship between the peak current and the square root of scanning rates.As shown in Fig.4c,the slope of calibration curves represent the effective surface area of the corresponding modified electrode.After the succesive depositions of Pt-CNF and PB step by step (Table 1), the effective surface area enormously increased from 0.032 cm2(Au electrode)to(0.080±0.004)cm2(30-PB/Pt-CNF).The higher effective area and lower Rctof the composite film may be resulted from the contribution of the Pt-CNF skeleton, which can exert a number of effects: (1) the provision of a large surface area that promotes the nucleation and growth of PBNPs,(2)the formation of an interconnected pathways enable to accelerate the electron transfer through the Pt-CNFs,and(3) the PBNPs on each Pt-CNF, minimizing barriers both to electrons transfer across the PB/electrolyte interface and electrons transport within the PB phase.

Table 1 Comparison of electrochemical effective surface area of different modified electrode

As verified before, the assembly cycles played a critical role on the electrochemical behaviors of the obtained PB/Pt-CNF films.Therefore, a series of PB/Pt-CNF films assembled with various cycles were prepared in this work.As shown in Fig.2d-g, the FESEM images showed the morphology evolutions of the PB/Pt-CNF films with increasing the assembly cycles.It can be found that more regular PBNPs were produced on the 3D Pt-CNF skeletons from 10 to 30 cycles.Meanwhile, PBNPs with more available surface area were produced on the Pt-CNF skeletons (Fig.4e and Table 1), as well as the obviously increased peak currents in CV curves (Fig.4d).However, as abundant PBNPs were created on the skeleton with 40 cycles,the intergrowth of PB crystals become more serious, forming a close-packed structure to cover the cavities produced by the random stack of CNFs.In this case,the laccase and detection target will be difficult to permeate into the inner space of this film,which causing the evident decrease of the active sites for the enzymatic reaction.According to the results in the Randles’s slope of Table 2,the effective area of 40 cycles was lower than that of 30 cycles to agree with the features in Fig.2f and g.Besides,gradually increased peak potential difference(ΔEp)values were observed with more assembly cycles (from 10 cycles to 40 cycles,the ΔEpis 54,66,73 and 86 mV,respectively),revealing that the electron transfer was hindered due to the increasing loading amount of PBNPs.

Table 2 Performance comparisons of reported HQ biosensors in the literatures

The surface coverage of the PBNPs on the 3D skeletons was evaluated according to the Faraday’s law [41,42]:

where Q is the total charge of a single peak,n is the average electron transfer (calculated by 57/ΔE), F is the Faraday constant (96485.34 C·mol-1) and A represents the electrode area (0.032 cm2).The surface coverage was estimated as 5.43, 9.94, 15.21 and 15.96 nmol·cm-2to the assembly cycles from 10 to 40 cycles.This result indicates that the surface coverage of 30 cycles assembled PB/Pt-CNF nearly reach saturation.Evidently, the morphologies of PBNPs on the 3D Pt-CNF skeletons could be rationally controlled through adjusting the assembly cycles, providing an effective approach to optimize the sensing performance of the prepared films.

In addition, the stability of the PB/Pt-CNF film modified electrode was examined by repetitive CV scanning for 30 times.A stable peak current with a subtle decrease rate of 7.32% was observed (Fig.4f), indicating a satisfactory stability was obtained on the PB/Pt-CNF film.The fabricated 3D PB/Pt-CNF film possessed a high electrocatalytic activity,a good stability and biocompatibility, make it a promising candidate in the construction of sensitive and stable enzymatic biosensors.

3.2.Performance of the PB/Pt-CNF based hydroquinone(HQ)biosensor

For constructing a HQ biosensor,screen-printed electrode(SPE)was applied to replace Au electrode.The SPE was modified by PB/Pt-CNF according to the experimental section.A laccase layer was immobilized on the optimum PB/Pt-CNF electrode through crosslinking.The electrochemical behaviors of the as-prepared biosensor were also investigated by CVs.As shown in Fig.4a and b, an obvious decline in the peak currents and increase in ΔEpwas observed after the enzyme loading.Meanwhile, a significantly enlarged semicircle in EIS test was acquired, indicating a higher Rctwas generated.Above results are ascribed to the poor conductivity of the laccase layer which severely impeded the electron transfer across the electrode surface.To investigate the influences of laccase and other phenolic components during the electrochemical process, we tested the biosensor in different conditions, as shown in Fig.5a-c.In the absence of laccase, both SPE and PB/Pt-CNF modified SPE were unable to obtain an obvious oxidation current change after the introduction of HQ, indicating that the SPE provided a weak oxidation to HQ.With the immobilization of laccase, a couple of well-defined redox peaks were observed and slightly changed after the addition of 50 μmol·L-1PE and RC.In contrast, when injecting 50 μmol·L-1HQ in the detection system,a stronger peak current at 0.2 V was obtained and almost unchanged after introducing CC at the same concentration, illustrating that this biosensor is much more sensitive to HQ than other phenolics.

Fig.5.(a) CVs of bare SPE without laccase immobilization in PBS and 50 μmol·L-1 HQ; (b) CVs of PB/Pt-CNF modified SPE without laccase immobilization in PBS and 50 μmol·L-1 HQ;(c)CVs of PB/Pt-CNF modified SPE with laccase immobilization in PBS and 50 μmol·L-1 phenolic solutions.(d)Amperometric responses of HQ biosensor to successive injections of HQ in PBS solution at 0.15 V;(e)The corresponding calibration curves for HQ detection;(f)Amperometric responses of CC to successive injections of CC in PBS solution at 0.25 V.

Then the continuously electrocatalytic performance of the asprepared HQ biosensor was tested by the chronoamperometry,and the working potential was chosen as 0.15 V.As shown in Fig.5d, the current responses were online recorded with the successive injections of different HQ concentration.The current responses were immediately produced following each HQ addition,and sensitive and steady steps were even observed up to the addition of 1.4 mmol·L-1HQ.The linear calibration of the as-prepared biosensor was fitted,in which a sensitivity of 220.28 μA·L·mmol-1-·cm-2was obtained (Fig.5e) with a wide linear range from 2.5 to 1450 μmol·L-1.This performance was further compared with those of the reported biosensors in literatures (Table 2).Among these,our prepared biosensor based on carbon ink substrate shows a widest detection range from 2.5 to 1450 μmol·L-1as well as a low working potential,benefiting the monitoring of HQ at a higher concentration.In addition, this SPE based biosensor integrated three electrodes on one chip, which simplified the operation process and was more competitive in real samples detection.

Table 3 Detection results for HQ in real water samples

As a comparison,similar test on the bare SPE electrode was also performed (Fig.5d), and significantly reduced current responses were acquired in this condition.As a result,a much lower sensitivity of 95.86 μA·L·mmol-1·cm-2was obtained(Fig.5e).It is revealed that the in-situ constructed 3D Pt/CNF skeleton significantly improved the sensing performance in the HQ detection, due to its large specific area and excellent electrical conductivity.

In addition, to verify the selectivity of this biosensor to HQ, the working potential was set as 0.25 V and CC was continuously added into the solution.As a result, there was no obvious current response(Fig.5f), indicating that this biosensor enables to specifically recognize HQ.

3.3.Anti-interference, reproducibility and stability of the biosensor

The electrochemical stability of the biosensor is critical for practical applications.As shown in Fig.6a-c,after 50 times of CV scanning cycles in PBS buffer, PBS with 1 ml real water and real water system, both oxidation and reduction peaks exhibited slight change.These results showed that the PB/Pt-CNF based biosensor possesses excellent electrochemical redox stability in the real water detection process.Meanwhile, the reproducibility of the biosensors was evaluated by using 10 independent SPEs (Fig.6b).The relative standard deviation (RSD) of the obtained sensitivities is estimated to be around 7.71%, indicating that the prepared HQ biosensors possessed satisfactory reproducibility.

Fig.6.(a)Stability of the biosensor after scanning in PBS with 50 μmol·L-1 HQ for 50 cycles;(b) Stability of the biosensor after scanning in PBS with 1 ml real water for 50 cycles;(c) Stability of the biosensor after scanning in real water for 50 cycle;(d)Reproducibility of the biosensor in 50 μmol·L-1 HQ solution;(e) Selectivity tests of the asprepared HQ biosensor at 0.15 V in 0.05 mol·L-1 PBS by alternately adding 20 μmol·L-1 HQ and 20 μmol·L-1 interferents every 20 s; (f) Stability of the sensor for 30 days.

As a common issue in the electrochemical sensing, the antiinterference ability of a biosensor is essential to realize an accurate assay in practice.Hence, the other 3 phenolic compounds (PE, RC and CC), fructose (Fru), sucrose (Suc), glutamate (Glu), bovine serum albumin (BSA), uric acid (UA) and ascorbic acid (AA) were introduced to investigate the selectivity of the prepared biosensor.As suggested in Fig.6c, the current responses showed no obvious change in the presence of other interferes, showing an excellent anti-interference ability of the prepared biosensors.

Moreover, the laccase immobilized PB/Pt-CNF electrode was stored in the refrigerator at 4°C for one month and then examined every 5 days after adding the HQ.As enhanced interactions of laccase on the composite film was formed due to the channel structures on the prepared film, the experimental results (Fig.6d)showed that the fabricated biosensors retained ～90.2%of their initial sensitivities, indicating an excellent stability of the fabricated biosensors.

3.4.Practical application of the HQ biosensor in real water samples

In order to evaluate the possibility of the PB/Pt-CNF based HQ biosensor in the analysis of real samples, the biosensor was employed to measure HQ in real water samples, including tap water and lake water from Jinghu Lake on our campus.Prior to the test, the water samples were filtered by PES microfiltration membranes (0.22 μm) to remove impurities.The results were shown in Table 3, in which the corresponding detection results were observed compared to the added amounts, revealing that the prepared biosensor has a great potential in the assay of real samples.

4.Conclusions

In this work, a novel 3D PB/Pt-CNF composite film was constructed to precisely recognize HQ in water.Due to the synergistic effects,the prepared film possessed a large specific area,high conductivity and excellent electrocatalytic activity.As a selected example, a high-performance HQ biosensor was constructed on the film, which exhibited a high sensitivity of ～220.28 μA·L·mmol-1·cm-2with a wide linear range from 2.5 to 1450 μmol·L-1, an excellent stability, and a high accurate assays of real water samples.It is highly anticipated that the prepared 3D PB/Pt-CNF film will have wide applications in the construction of more high-performance electrochemical enzyme biosensors.

Declaration of Competing Interest

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.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China(22078148 and 21727818),the Innovative Research Team Program by the Ministry of Education of China (IRT_17R54), the Top-notch Academic Programs Project of Jiangsu Higher Education Institutions(TAPP),the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD),the Key Project by Medical Science and Technology Development Foundation of Nanjing Department of Health (ZKX17014)and the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX20_1021).