Longqi Ling,Jio Yin,Jinpeng Bo,Linhun Cong,Weimin Hung,*,Hio Lin,Zhn Shi

a College of Chemistry,Jilin University,Changchun 130012,China

b Key Laboratory of Functional Materials and Devices for Special Environments,Xinjiang Technical Institute of Physics&Chemistry,Chinese Academy of Sciences,Urumqi 830011,China

c Guangdong Guanghua Sci-Tech Co.,Ltd.,Shantou 515061,China

d College of Chemistry,State Key Laboratory Inorganic Synthesis&Preparative Chemistry,Jilin University, Changchun 130012,China

Key words:TiO2NA Au NP Electrodeposition COD Photoelectrochemical catalysis

ABSTRACT Au nanoparticles(Au NPs)were electrodeposited at the highly ordered anatase TiO2 nanotube array(TiO2NA)electrode under sonicating,and the Au NP-TiO2NA sensor was characterized by scanning electron microscope(SEM),X-ray photoelectron spectroscopy(XPS),X-ray diffraction(XRD).The photoelectrochemical experiments indicate the Au NP-TiO2NA sensor has lower photoelectro-resistance,higher photoelectrocatalytical activity and stability than that of pure TiO2NA sensor under the same conditions.The as-prepared sensor can be used for the determination of chemical oxygen demand(COD)in real samples,and the obtained resultsare comparable well with those of by standard K2Cr2O7 method.The method proposed is simple,fast,cost effective,and environmentally friendly.

Chemical oxygen demand(COD)is one of the most important parameters and has been w idely employed for w ater quality assessment.The standard method for COD determination,the K2Cr2O7method,requires re fl ux over a long period of time under high temperature to achieve adequate oxidation and also resultsin the consumption of expensive(Ag2SO4),corrosive(H2SO4),and toxic(Hg2+and Cr2O72?)chemicals[1].In an attempt to shorten the time required for analysis,modified K2Cr2O7methods have been developed based on microw ave-assisted oxidation[2]or ultrasound-assisted oxidation[3].However,the secondary pollution is unavoidable w hen the standard method is employed.

Recently,great efforts have been devoted to the development of rapid,accurate,and environmentally friendly methods for the determination of accurate COD values[4–12],such as electrochemical methods using nano-PbO2modified electrode[4],borondoped diamond sensor[5],or composite planar electrode[6],and photocatalytic and photoelectrocatalytic methods based on TiO2nanomaterial sensors[7–12].However,these methods cannot be satis fied.For example,the nano-PbO2modified electrodespose the risk of the potential release of hazardous Pb during the preparation[4].Extensive research has show n that TiO2isan excellent material for photodegradation of organic pollutants in w ater.The TiO2nanomaterial is non-toxic,inexpensive,photosensitive,photostable,and environmentally friendly.

The main disadvantage of TiO2-based photocatalytic methods in practical applications is the easy recombination of the photogenerated electron/hole pairs in discrete TiO2nanoparticles and coated nano fi lms,which results in lower photocatalytic activity.For the determination of CODvalue,this implies a narrow dynamic working range and relatively poor reproducibility[12].Compared with traditional analytical methods,the photoelectrocatalytic degradation approach is more effective because of the suppression of photohole and photoelectron recombination.However,the preparation processes of TiO2nanoparticle by sol-gel method are time consuming under high pressure and temperature[7–10].And the structural disorders may hinder efficient electron transport[13].Recently great interesting haspaid to the investigation of TiO2nanotube array(TiO2NA)electrode[11,12,14].The ordered architecture of TiO2NA can provide a unidirectional electronic channel and reduce the grain boundaries.The TiO2NA show s a stronger attachment to the parent titanium substrate and a better photoelectrocatalytic activity ow ing to the improved electron transport and reduced charge recombination.Nevertheless,the TiO2NA sensors show unsatisfactory photocatalytic activity,shorter lifetime,and inferior stability in practical application for COD assay.

Consequently,it is necessary to develop new methods to compensate the shortcomings of present TiO2sensors.Metal-and nonmetal-doping of TiO2have been proved to be effective ways to enhance photocatalytic activity[15].For example,it was reported that Au/TiO2nanocomposites or TiO2-supported Au nanoparticles(AuNPs)enhanced photocatalytic activity by embedded Au NPsinto TiO2gel-sol or nano fi lms[16].For example,Au NPs embedded within the framework of TiO2may enhance light absorption and improve quantum ef fi ciency during the photocatalytic processes[16].Therefore,we try to modify the TiO2NA by AuNPs(Au NPTiO2NA)to improve the sensitivity,stability,and lifetime of TiO2NA sensor.The as-prepared sensor was characterized by scanning electron microscope(SEM),X-ray photoelectron spectroscopy(XPS),X-ray diffraction(XRD),and electrochemical impedance spectroscopy(EIS).The performance of pure TiO2NA and Au NPTiO2NA sensor for COD determination was also compared.

In this paper,we conducted the experiment to prepare the TiO2NA and AuNP-TiO2NA sensor by the follow ing methods.The 0.25-mm-thick Ti foil(1 cm?5 cm)is anodized at 20 V in 0.5 w t%HF solution,according to the method proposed by Liu et al.The voltage is applied by a DCpower supply(Model SK1730SL,Sanke,China).Prior to anodization,the Ti foil is chemically etched by immersing in a mixture of HFand HNO3for 30 s(the mixture ratio of HF:HNO3:H2O is 1:4:5 in volume)followed by sonicating in acetone,ethanol,and w ater for 5 min,respectively.The crystallization of TiO2is realized by annealing the as-prepared foil at 500?C under ambient air for 3 h.Au NP-TiO2NA sensor is prepared by electrodeposition of Au NPs at the TiO2NA in 0.05 mol/L H2SO4+5.0 mmol/L HAu Cl4solution at 0 V under sonicating.

The SEM(JEOL Ltd.,Japan)is used to characterize the TiO2NA and Au NP-TiO2NA fi lm surface.Fig.1 show s the SEM images of the highly ordered TiO2NA fi lm with and without Au NPs.It is observed that high-density,well-ordered,and uniform TiO2NA is synthesized by the electrochemical anodic oxidation of pure titanium foil.The tops of the tubes are open,the diameters of these nanotubes range from 60 nm to 80 nm with w all thickness of about 16 nm and length of 350–380 nm.After electrodeposition under sonication,the AuNPs with average diameter(3?2)nm are uniformly distributed in TiO2nanotubes or w alls,as show n in Fig.1c.Moreover,the surface of the the TiO2NAand Au NP-TiO2NA fi lm was translated into 3D frameworks as show n inFigs.1d–f,which promote the discussion furthermore.

We further analyzed Au NP-TiO2NA fi lm surface composition using XPS.It is well know n that XPS is a valuable technique for detecting the surface composition of samples.XPSmeasurement is performed on an ESCALAB-MKIIspectrometer(VGCo.,UK)with an Al K a X-ray radiation as the X-ray source for excitation and a chamber pressure of 3.5?10?7Pa.Fig.S1(Supporting information)presents the XPS spectra of AuNP-TiO2NA Ti 2p,O 1s,Au 4f.In Fig.S1a,the tw o peaks at 458.7 eV and 464.1 eV are assigned to the Ti 2p3/2and Ti 2p1/2states in TiO2,respectively.The doublet peaks are due to the spin-orbit splitting of Ti 2p and separated by 5.4 eV.Both of the peaksare in good agreement with those of pure anatase TiO2.The peak at binding energy of 529.95 eV corresponds to bulk oxygen bonded to titanium.In Fig.S1c,the tw o peaks at 84.0 eV and 87.7 eV can be assigned to Au 4f7/2and Au 4f5/2.The XPSresult indicates that the Au species are present in the metallic state.

Moreover,we studied the crystalline nature of the Au NPTiO2NA surface using XRD(Philips X-ray diffractometer,model PW1700 BASED,Cu K a radiation l=1.5406?),as show n in Fig.S1d.The diffraction peaks at about 2u=25.4?,48.1?can be indexed to the(101)and(200)crystal faces of anatase TiO2.The other diffraction peaksat about 2u=38.5?and 40.3?can be indexed to the titanium substrate.The peak at 2u=44.5?is assigned to the(200)plane of electrodeposited AuNPs at TiO2NA surface[16].The diameter of AuNPs can be calculated according to the Scherrer equation:d(?)=kl/b cos u,w here k is a coefficient(0.9),l=the w avelength of X-ray used(1.54 nm),b=the full-w idth halfmaximum of respective diffraction peak(rad),u=the angle at the position of peak maximum(rad).The diameter of Au NPs is about 3 nm according to the Scherrer’s equation,which is consistent with the result obtained by the SEM image.

Fig.2.Nyquist diagrams plots(Zim vs.Zre)for TiO2NA(a)and Au NP-TiO2NA(b)sensor in the presence of 0.10 mol/Lferri/ferrocyanide redox couple in a 0.10 mol/L KCl solution with UV illumination under different bias potentials.

After understanding its basic appearance and composition,we conducted photoelectrochemical test.An appropriate positive bias potential to the working electrode plays an important role in photoelectrocatalytic reactivity.Fig.2 compares EISof the TiO2NA sensor with and without AuNPs.EISmeasurements are carried out with an Autolab/PG30 electrochemical analyzer system(ECO Chemie B.V.,Netherlands)in a grounded Faraday cage.The Nyquist plots show the TiO2NAand AuNP-TiO2NAsensor in the presence of 0.10 mol/L ferri/ferrocyanide redox couple in a 0.10 mol/L KCl solution under different bias potentials with UV illumination.The plots reveal that Rct are increased as the applied potential increased.For example,Rct of the TiO2NA sensor at the anodic potential of 0.05 V,0.3 V,and 0.6 Vare ca.22.41,28.43,and 40.9 k V,respectively.The diagrams of Au NP-TiO2NA sensor are similar,including a semicircle part and a straight line part.The semicircle at high frequencies is characteristic of the charge transfer process and the linear part at low frequencies corresponded to the diffusion-controlled step[17].The Rct of the Au NP-TiO2NA sensor are also increased from ca.1.59 k V(at 0.05 V)to 4.66 k V(at 0.3 V),and to 9.0 k V(at 0.6 V).The Au NP-TiO2NA sensor exhibited much lower resistance than that of TiO2NA sensor under the same conditions,which can be attributed to the deposited Au NPs could improve quantum ef fi ciency and the charge transfer processes of TiO2NA sensor.Meanwhile,Au NPs embedded within the framework may also serve as an electric conductor,which facilitates photoelectron transfer to pore surface and further reduce the probability of charge recombination.Consequently,the recombination center of photogenerated carriers was diminished and the separation effect of them was improved.As is show n in Fig.2,the bias potential of 0.05 V is enough to separate photoelectrons and photoholes.Such a low potential could diminish side reaction during the photoelectrocatalysis such as photolysis of H2O.Therefore,0.05 V is selected for COD determination.

Finally,a series of standard glucose solutions of different COD values were checked with the AuNP-TiO2NA sensor(Fig.3a).The photocurrent was linear in relation to the COD value in the range from 5 mg/L to 100 mg/L. The calibration curve was y=7.728+0.8010x,R=0.9945,w here x and y were the values of COD(mg/L),as show n in Fig.3a1.The detection limit was 5 mg/L based on three times the standard deviation of the baseline.The Au NP-TiO2NA sensor exhibits lower background photocurrent,higher photoelectrocatalytic activity,and w ider linear range than those of TiO2NA sensor under same conditions.The photocurrent of Au NP-TiO2NA(Fig.3b,curve 1)is higher than those of pure TiO2NA electrode(Fig.3b,curve 2)tow ard the oxidation of standard glucose solutions.To compare the stability of the TiO2NA and Au NP-TiO2NA sensor,ten successive photocurrent data with and without organics were achieved.The relative standard deviations(RSDs)were 3.90%and 0.50%for TiO2NA and Au NPTiO2NA sensor without any organics in the solution,and the RSDs were 4.00%and 0.61%for TiO2NA and Au NP-TiO2NA sensor in the presence of 5 mg/L O2of glucose,respectively.The AuNP-TiO2NA sensor exhibited better stability under the same conditions.The long-term stability of the AuNP-TiO2NA sensor was tested over a 30-day period.During this period,the slope change of calibration curve was5.60%.The RSDobtained for successive measurements of 20 mg/L O2glucose solution was 7.81%.Excellent stability may be attributed to strong attachment between deposited Au NPs and TiO2NA of as-prepared sensor.

In the Meanwhile,we also tested the actual wastew ater samples.Wastew ater samples are collected from Southlake of Changchun,wastew ater treatment plants,and food factory according to the guidelines of the standard method[18].Standard addition method is used to determine the COD value of the samples.The results are also compared with the standard K2Cr2O7method.The result correlation between the tw o methods is show n in Fig.3c.A highly signi fi cant correlation(y=2.916x?2.666,R=0.9890,P<0.0001,n=10)between the tw o methods was obtained,w here x and y are the results obtained by using K2Cr2O7method and the proposed method,respectively.The RSD obtained from the tw o methods for CODs are less than 5%,the results indicate that the tw o methods are compatible well.The proposed method is valid for the determination of COD in real samples,though the COD values of real samples determined by Au NPTiO2NA sensor are about three times(2.916)higher than of the standard K2Cr2O7method.

In summary,Au nanoparticles modified TiO2nanotube array electrode was facilely prepared via simple electrodeposition process,offering more probability to regulate the morphology and constituent,as show n in Fig.4.After applying as chemical oxygen demand(COD)sensor,the unique electrode show seffective impact in waste w ater.The method proposed isso simple,fast,cost effective,and environmental friendly that w ill provide train of thought for further research.

Acknow ledgm ents

This work was supported by the National Key Research and Development Program of China(No.2016YFC1102802)and Guangdong Innovative and Entrepreneurial Research Team Program(No.2013C092).

Fig.3.The calibration curve of photocurrent obtained from glucose standard solutions of different COD values at AuNP-TiO2NA(a1).Photocurrent obtained from glucose standard solutionsof different CODvalues at Au NP-TiO2NA(a)and comparison of TiO2NAand Au NP-TiO2NA(b)sensors,respectively.(c)The result correlation between the asproposed sensor method and the standard dichromate method for different real samples.1-3 are the samples collected from Southlake of Changchun,wastewater treatment plants,and food factory and diluted to different concentrations,respectively.

Fig.4.The sketch of w hole process from facile preparation to effective application.

App endix A.Supp lem entary data

Supplementary data associated with this article can be found,in the online version,at https://doi.org/10.1016/j.cclet.2018.01.049.