Novel helical TiO2 nanotube arrays modi

Novel helical TiO2nanotube arrays modi?ed by Cu2O for enzyme-free

glucose oxidation

Mei Long,Lin Tan,Hongtao Liu,Zhen He,Aidong Tang n

School of Chemistry and Chemical Engineering,Central South University,Changsha410083,PR China

a r t i c l e i n f o

Article history:

Received29January2014

Received in revised form

8March2014

Accepted12March2014

Available online29March2014

Keywords:

Cu2O/TiO2

Nanotube array

Electrochemical property

Sensor

a b s t r a c t

A helical TiO2nanotube(TNT)array modi?ed with cuprous oxide(Cu2O)electrode was fabricated and

used for nonenzymatic glucose detection.The structure and morphology of Cu2O/TNT were character-

ized by X-ray diffraction and transmission electron microscopy.The electrocatalytic performance of

Cu2O/TNT electrode for glucose oxidation was investigated by cyclic voltammetry and chronoampero-

metry.At an applied potential oft0.65V versus SCE,a linear range was obtained within the

concentration range of3.0–9.0mM with a detection limit of62μM(signal/noise?3).The response

time was approximately3s after adding0.10mM glucose.Formate and gluconic acid were identi?ed as

the main products of the glucose oxidation using1H NMR spectrometry.A possible mechanism for

continuous glucose oxidation was also proposed.

&2014Elsevier B.V.All rights reserved.

1.Introduction

Glucose is a simple sugar that is extensively distributed in the

blood of living organisms.This simple sugar is often used as a

marker of diabetes,which has become one of the major diseases

worldwide.Moreover,glucose as a common food component is

also an indicator of food quality/acceptability.Therefore,the

quantitative determination of glucose level,not only in blood but

also in other sources such as foods and pharmaceuticals,is very

important in biological and clinical analysis(Chen,2013;Wang et

al.,2012).Glucose oxidase exhibits high sensitivity and selectivity

to glucose and is widely used as biosensors for glucose detection.

However,enzymatic glucose sensors are limited by their instabil-

ity,complex enzyme immobilization,and high sensitivity to

temperature,pH,and humidity.To overcome these drawbacks,

numerous efforts have been made to develop non-enzymatic

amperometric glucose sensors.These sensors are based on the

generation of electrons in catalyst-assisted glucose oxidation,

resulting in a detectable current response that is linearly corre-

lated with glucose concentration(Guo and Wang,2011).With

ongoing technical development,nanostructured metals,such as Pt

(Park et al.,2012),Au(Wang,et al.,2013a,2013b),Ag(Yang,et al.,

2013),and Ni(Pissinis et al.,2013),or metal-oxides such as NiO

(Cao et al.,2011;Zhu et al.,2013),Co3O4(Hou et al.,2012;Ensa?

et al.,2013),and Cu x O(Jiang and Zhang,2010;Wang et al.,2010a,

2010b;Luo et al.,2011;Yang et al.,2012;Zhang et al.,2012;

Dung et al.,2013;Sun et al.,2013)have been extensively explored

in constructing biosensors for the determination of glucose via

electrochemical oxidation(Rahman et al.2010).Compared to

noble metals and alloys,copper-based materials that are less

expensive exhibited comparable catalytic oxidation ability for

glucose(Zhang et al.,2009;Li et al.,2010;El Khatib and

Hameed,2011;Luo et al.,2012;Wang et al.,2012;Yec and Zeng,

2012;Zhou et al.,2012;Meng et al.,2013;Xu et al.,2013).One

promising strategy for Cu2O,used as a high-ef?ciency electro-

catalyst,is planting Cu2O nanoparticles to various supporting

materials such as polypyrrole(PPy)nanowire(Meng et al.,2013),

reduced graphene oxide(Zhang et al.,2009;Xu et al.,2013),and

straight multi-walled carbon nanotube(Zhou et al.,2012).The

Cu x O/PPy sensor showed excellent selectivity,reproducibility,and

stability(Meng et al.,2013).Furthermore,cuprous oxide-reduced

graphene oxide nanocomposite displayed better performance for

the catalytic reduction of H2O2,compared with the single compo-

nent Cu2O(Xu et al.,2013).Similarly,the as-prepared nanospin-

dle-like Cu2O/SMWNTs nanohybrids demonstrated much higher

electrocatalytic activity towards the oxidation of glucose than the

SMWNTs or Cu2O alone,which is attributed to the high catalytic

active sites provided by the nanospindle-like Cu2O,and the high

electron transfer rate delivered by an ef?cient electrical network

formed by Cu2O and SMWNTs(Zhou et al.,2012).Recently,the

encapsulation of individual nanoparticles in morphologically well-

de?ned inorganic tubes has received much attention because of

thermal and chemical stabilities in catalysis(Huo and Yang,2012).

The substrates of nonenzymatic sensors and their spatial structure

have been recognized as the main factors that in?uence the

analytical performance of biosensors(Meng et al.,2013).

Contents lists available at ScienceDirect

journal homepage:https://www.360docs.net/doc/fb416563.html,/locate/bios

Biosensors and Bioelectronics

https://www.360docs.net/doc/fb416563.html,/10.1016/j.bios.2014.03.032

0956-5663/&2014Elsevier B.V.All rights

reserved.

n Corresponding author.Tel./fax:t8673188879616.

E-mail address:tangaidong@https://www.360docs.net/doc/fb416563.html,(A.Tang).

Biosensors and Bioelectronics59(2014)243–250

Well-aligned TiO2nanotube arrays(TNT)with a large speci?c surface area and abundant active sites have attracted extensive interest because of their low cost,long-term chemical and thermal stability,excellent biocompatibility,and nontoxicity.Consequently, the TNT has been proposed as one of the ideal carriers for the immobilization of biomolecules(Rahman et al.,2010),metals (Wang,2010a,2010b;Li et al.,2013),and metal oxides.However, compared to electron conductors,such as PPy nanowire,reduced graphene oxide,and straight multi-walled carbon nanotube,TiO2 has very poor conductivity.We proposed that Cu2O in tandem with TiO2would be a good solution for this problem.The narrow-bandgap Cu2O nanoparticles,acting as active materials to promote the charge transfer to TiO2/Ti electrode,lead to an ef?cient charge carrier separation,thus improving the conductivity of TiO2/Ti electrode.Therefore,such p–n heterojunction electrodes com-posed of Cu2O-loaded TiO2nanotube arrays may exhibit enhanced electrocatalytic activities.Various methods have been adopted to obtain catalysts of different morphologies with high surface area. Previous studies on Cu2O/TiO2heterojunction nanoparticles have been reported(Huang et al.,2010;Zhang et al.,2011;Chen,2013; Wang,et al.,2013a,2013b).Electrochemical deposition is one of the most convenient and effective methods to deposit semicon-ductor nanoparticles onto the TNT structure.Recently,anodic TiO2 nanotubular arrays having high surface area and precisely con-trolled morphology have been widely studied as catalyst support. However,to the best of our knowledge,the helical TNTs are less studied,and no reports on Cu2O/TNT array for nonenzymatic glucose detection have been published.In this work,we success-fully fabricated helical Cu2O/TNT composites by incorporating the nano-sized polyhedral Cu2O into the entire TNT frameworks using an electrodeposition method.The synergy between TiO2nano-tubes and Cu2O nanopaticles may lead to an enhanced detection performance of nonenzymatic glucose sensors.The activity of Cu2O/TNT nanostructures may also be improved because of high catalytic active sites provided by Cu2O nanoparticles supported on the self-assembled and highly oriented TiO2nanotube arrays with a helical-like morphology.The electrochemical behavior of the obtained anodic curves toward electrocatalytic oxidation of glu-cose,including a proposal for the detection mechanism,is also discussed.

2.Experimental

2.1.Chemicals

All chemicals used in this study were of analytical grade. Glucose,CuSO4á5H2O,NaOH,NH4F,glycerol,lactic acid,and Ti foils(purity499.6%)were purchased from ChangSha Chemical Corp.in China.Glucose was used as received.Double-distilled water was used for the preparation of standard solutions.

2.2.Preparation of the Cu2O/TNT electrode

TNTs were fabricated through a two-step anodization.Before fabrication,Ti foils(purity499.6%,100mm?10mm in size, 0.14mm in thickness)were ultrasonically cleaned in acetone and alcohol.A Pt foil(purity499.9%)was used as the cathode and pure Ti foil was used as the anode.In the?rst step,anodization was carried out in a glycerol electrolyte containing0.5wt%NH4F. The voltage and oxidation time were20V and1h,respectively. Afterward,the pre-anodized Ti foil was ultrasonically cleaned in an acid solution containing HF and HNO3to remove the formed TiO2?lm.The cleaned Ti foil was used for the secondary anodiza-tion in a similar electrolyte with the addition of HF.The voltage and oxidation time were20V and4h,respectively.The as-prepared TiO2nanotubes were cleaned using distilled water, air dried,and then calcined in a muf?e furnace at5001C for3h.

Cu2O/TNT composites were prepared by electrochemical deposition of Cu2O in a three-electrode cell,using the TNT, saturated Hg/Hg2Cl2(SCE),and platinum foil as the working, reference,and counter-electrodes,respectively.A CHI660B elec-trochemical working station(ChenHua Instruments Co.Ltd., Shanghai,China)was used as the power source.The electrolyte was obtained by dissolving1mol Là1CuSO4in a3mol Là1lactic acid solution to form a copper lactate complex,and pH was further adjusted to12using5M NaOH solution.The electrolyte was stirred and kept at a constant temperature of351C.An electro-deposition charge of1C and a potential ofà0.7V versus saturated calomel electrode were used during the deposition process.After the electrochemical deposition,the Ti foil covered with Cu2O/TNT was thoroughly washed using distilled water,and air dried.

2.3.Characterization

X-ray diffraction(XRD)patterns of the fabricated products were recorded on a DX-2700X-ray diffractometer(Liaoning Dandong Hao Yuan Instrument Co.,Ltd.TD)at a scan rate of 0.051sà1with a2θrange from201to801,using high-intensity Cu Kαradiation(λ?0.15418nm).Scanning electron microscope

(SEM)images were obtained using a FEI Nova NanoSEM230?eld emission scanning electron microscope.Microstructures and morphologies were investigated using a Tecnai G2F20S-TWIN (FEI Company)transmission electron microscope(TEM)with a ?eld emission gun of200kV.1H NMR spectra were recorded using an AVANCE III500MHz Bruker spectrometer.Tetramethylsilane (TMS)was used as the internal standard(0.00ppm)and the residual solvent was D2O.

2.4.Electrochemical measurements

Electrochemical measurements were performed using a CHI 660B electrochemical workstation(ChenHua Instruments Co.Ltd., Shanghai,China)on a conventional three-electrode setup,with Cu2O/TNT as the working electrode,Pt foil as the counter-elec-trode,and SCE as the reference https://www.360docs.net/doc/fb416563.html,ing the modi?ed Cu2O/TNT working electrode,data on cyclic voltammetric(CV) and amperometric experiments were obtained in a mixture of 4mmol Là1glucose and0.1mol Là1NaOH.The CV measurements required operation of the electrode in a range of potential of 0–0.8V versus SCE.The measurements of amperometric curves required operation of the electrode at a constant applied potential of0.65V versus SCE under ambient temperature.Once the current reached a plateau in the absence of glucose,glucose was added every50s thereafter with constant stirring using a magnetic stirrer at20rpm.

3.Results and discussion

3.1.Characterization of the Cu2O modi?ed TiO2nanotubes array electrode

The prepared TiO2nanotubes array electrode was calcined at 5001C for3h;this pre-annealing treatment of the TiO2nanotube arrays is bene?cial to the deposition of Cu2O into TiO2tubular structure(Huang,Peng et al.,2011).Fig.1shows the XRD patterns of the annealed bare TNT and Cu2O modi?ed TNT.

The peaks at2θ?25.31and48.11in Fig.1a are attributed to the diffraction peaks of anatase TiO2(JCPDS?le:21-1272).The inter-planar distance of(101)crystal planes of anatase TiO2was measured to be0.3504nm and the crystal size was calculated to

M.Long et al./Biosensors and Bioelectronics59(2014)243–250 244

be about 34.18nm by the Scherrer equation (Yang et al.,2011).Three kinds of diffraction peaks can be identi ?ed from the XRD patterns in Fig.1b.The presence of Cu 2O in the sample after deposition can be con ?rmed by the characteristic re ?ection peaks at 36.51and 29.61,assigned to (111)and (110)crystal planes of the cubic Cu 2O (JCPDS ?le:65-3288).The interplanar distances of (111)and (110)crystal plane of cubic Cu 2O were measured to be 0.2456nm and 0.2978nm,and the crystal size of (111)crystal planes was calculated to be about 40.95nm by the Scherrer equation (Yang et al.,2011).Aside from the diffraction peaks of Ti substrate and TiO 2nanotubes,no other peak assigned to CuO or Cu phase was detected in the as-prepared sample.

Fig.2shows the morphologies and the crystal of the obtained Cu 2O/TNT.The top-view pro ?le-sectional FE-SEM images of the as-prepared TNT with a helical nanotube structure are shown in Fig.2a.Fig.2b illustrates the pro ?le-sectional TEM images of the Cu 2O/TNT.The Cu 2O modi ?ed TNT shows a highly ordered and open structure with a clearly helical-like morphology.The classic Cu 2O/TNT arrays with an outer diameter of about 105nm and wall thickness of 10nm in Fig.2c are highly arranged.

Fig.2d and e shows high resolution TEM (HRTEM)images of the as prepared Cu 2O/TNT.The observed lattice spacing of 0.35nm in the left portion of the image corresponds to (101)plane of anatase,showing that the wall consists of TiO 2.The observed 0.245and 0.296nm fringes of the wall correspond to (111)and (110)planes of Cu 2O,respectively.Results show that Cu 2O has been deposited into the outer pore wall of the TiO 2nanotubes.The small lattice distances difference between (110)planes of Cu 2O and (101)facets of TiO 2favors the (110)planes of Cu 2O nucleation on the surface of TiO 2nanotubes.HRTEM image of Cu 2O –TiO 2nanocomposites shows that the (110)planes of Cu 2O with d -spacing of about 0.296nm and the (101)facets of TiO 2with d -spacing of about 0.350nm intersect each other at the interfaces,which means that Cu 2O were epitaxially grown on the surfaces of TiO 2(Fig.2d).However,the large lattice mismatch between (111)planes of Cu 2O and (101)facets of TiO 2makes it very dif ?cult to form a desired heterogeneous nucleation on the TiO 2nanotubes surfaces (Fig.2e).Therefore,the interaction between TiO 2and Cu 2O is very likely attributed to the (110)facets Cu 2O epitaxially grown on the surfaces of TiO 2despite the necessity of further con ?rmation.TiO 2nanotube arrays clearly act as a support to facilitate the dispersion of Cu 2O nanoparticles.The TiO 2nanotubes have a well-ordered tubular structure with a high surface-to-volume ratio,which is bene ?cial for the attachment of Cu 2O nanoparticles and transport of electrical carriers along one con-trollable direction.Moreover,the helical tunnel structure and the

vertically aligned TiO 2nanotubes surrounded by numerous Cu 2O nanoparticles could improve the incorporation of analytes.3.2.Electrocatalytic oxidation of glucose on Cu 2O/TNT electrode The effect of pH on the glucose electrochemical property has been investigated on Cu 2O/MWCNTs modi ?ed electrode (Zhang et al.,2009).High pH is helpful to the mutarotation of glucose,which results in the increase of current.On the other hand,at high pH,glucose isomerization occurs and fructose and mannose will be formed,which results in the decrease of current.Therefore,we selected the NaOH concentration as 0.1M for nonenzymatic glucose detection.The electrocatalytic activities of bare TNT electrode and Cu 2O/TNT electrode towards the oxidation of glu-cose in 0.1M NaOH are shown in Fig.3A.In the presence of 4.0mM glucose,no obvious catalytic oxidation current on the bare TNT electrode (curve b)is observed compared with the absence of glucose (curve a).By contrast,the Cu 2O/TNT electrode shows obvious catalytic oxidation current (curve d)in the presence of 4.0mM glucose compared with the absence of glucose (curve c).The catalytic oxidation current is signi ?cantly higher than that of the Cu 2O/TNT electrode in the absence of glucose.The oxidation current rapidly increases,starting at about 0.35V with the appearance of an oxidation peak at around 0.55V and completing at about 0.75V.These results show that Cu 2O/TNT has good electrocatalytic activity toward the oxidation of glucose.Moreover,the catalytic currents gradually increase on the Cu 2O/TiO 2elec-trode with increasing glucose concentration (Fig.3B)and Fig.3C shows the cyclic voltammograms of the Cu 2O/TNT electrode recorded in 0.1M NaOH solution at different scan rates in the range of 20–200mV s à1.The oxidative peak current (the corre-sponding potential is 0.65V)is notably increased in a linear manner with increasing scan rates.

Fig.3C shows the cyclic voltammograms of the Cu 2O/TNT electrode recorded in 0.1M NaOH solution at different scan rates in the range of 20–200mV s à1.The oxidative peak current (the corresponding potential is 0.65V)linearly increased with the increasing scan rate in the range of 20–200mV s à1,as shown in the inset of Fig.3(C).A good linearity between scan rate and peak current is obtained,i (μAcm à2)??0.19t40.16ν(mVs à1),with a correlation coef ?cient (R 2)of 0.996.Further examinations about the dependence of scan rate on peak current at Cu 2O/TNT electrode illustrated a linear relationship between the voltam-metric peak current for glucose oxidation and the square root of the scan rate,àln i (μAcm à2)??1.51t0.026ν1/2(mV s à1)1/2,with a correlation coef ?cient (R 2)of 0.994.These results indicate that the glucose electrochemical oxidation is under diffusional control of glucose at scan rates lower than approximately 0.2V s à1.

To further clarify the effect of Cu 2O nanoparticles on electro-chemical properties of TiO 2nanotubes different Cu 2O/TNT electro-des were prepared at different deposition charges 1.0C and 1.5C,other conditions being the https://www.360docs.net/doc/fb416563.html,parison of their SEM images and CV curves is shown in Fig.3D.Different deposition charges would result in different amounts of Cu 2O nanoparticles being deposited onto TNT,which would generate different catalytic activities towards glucose oxidation at the prepared electrodes.As can be seen,Fig.3D(a,b)displays lots of Cu 2O particles with different morphologies decorated on the walls of TNT.Moreover,the response current of the Cu 2O/TNT electrode prepared at 1.5C is greater than that of 1.0C from Fig.3D(c,d),showing that more Cu 2O nanoparticles deposited onto TNT lead to higher response current.

To identify the products generated during glucose oxidation,1

H NMR spectrometry (Fig.4)was employed according to the method described by Farrell and Breslin (2004).For these mea-surements,glucose concentration was increased to 30

mM.

Fig.1.XRD patterns of TiO 2nanotube arrays (a)and Cu 2O/TiO 2nanotube arrays (b).

M.Long et al./Biosensors and Bioelectronics 59(2014)243–250245

The glucose solution was oxidized at an applied potential of 0.65V/(SCE)for a 16h period and then analyzed using TMS as the internal standard.

For the starting material,the expected absorptions for glucose in 2.8–3.8ppm region were demonstrated.The peak centered at 8.3ppm is attributed to formate and has been previously observed by Luo and Baldwin (1995).Absorption in the 3.75–4.75ppm region represents gluconic acid formation.Therefore,gluconic acid and formate were identi ?ed as the main products of the glucose oxidation reaction in 0.1M alkaline solution,detected using 1H NMR spectrometry.The obtained results indicate that the major products,gluconic acid and formate,are in agreement with the ?ndings of Marioli and Kuwana (1992),who concluded that glucose oxidation may proceed beyond the formation of gluconic or glucuronic acids to give lower molecular weight products and fragmentation of the carbohydrate carbon skeleton.3.3.Electrochemical sensing properties

Fig.5shows the amperometric measurement of the Cu 2O/TNT electrode in 0.1M NaOH solution at 0.65V versus SCE,with successive additions of glucose.The steady-state currents are obtained at a 3s response time and are plotted as the calibration curve,which shows the lowest limit of detection,62μM,for glucose.This observation may be attributed to the enhanced nano-size effect realized by the highly well dispersed self-assembly nanostructure.The sensor electrode exhibits linearity for glucose sensing that ranged from 3mM to 9mM,with a correlation coef ?cient of 0.992(illustrated by the right inset in Fig.5).The electrode sensitivity calculated from the slope of the calibration curve is 14.56μA cm à2mM à1.The Cu 2O/TNT electrode exhibits improved characteristics in terms of sensitivity and limits of detection,which is induced by the nano-size effect in a well dispersed self-assembly nanostructure.

As shown in Fig.6,to evaluate the selectivity of the Cu 2O/TNT electrode,the current responses to several possible interfering biomolecules,such as ascorbic acid (AA),uric acid (UA),and fructose,which normally coexist with glucose in human blood,were examined.Considering that the concentration of glucose in the human blood is more than 30times of interfering materials (Luo et al.,2011),the interference experiment was conducted at Cu 2O/TNT arrays electrode by successive addition of 4.0mM glucose and 0.40mM interfering materials into the 0.10M NaOH solution.The results show an apparent response to glucose and insubstantial responses to interfering materials.The interferences current responses range from 5.47%(AA)to 6.66%(fructose)and (UA)9.61%with respect to the current response to 4.0mM glucose at t0.650V (versus SCE).Hence it can be concluded that the Cu 2O/TNT electrode can demonstrate high selectivity for glucose detection.Although Cu-based electrode materials were used for very high potential in biosensor performance,the applied poten-tial of 0.65V for our obtained Cu 2O/TNT electrodes is not very high compared with the literature.For a glassy carbon electrode modi ?ed with CuO nanowires and a Na ?on ?lm,the

maximum

Fig.2.Top-view pro ?le-sectional FE-SEM images of the as-prepared TNT (a);the pro ?le-sectional TEM images of the Cu 2O/TNT (b);and the HRTEM images of the Cu 2O/TNT (c)–(e).

M.Long et al./Biosensors and Bioelectronics 59(2014)243–250

246

response current with a good signal/noise ratio was achieved at 0.80V(Zhang et al.,2012).In addition,for a copper nanocluster/ multiwall carbon nanotube-modi?ed glassy carbon electrode,the maximum response current was observed at0.65V and a good signal/noise ratio can be offered that is the same as our results. When the sensor was applied for the determination of glucose in real blood serum samples,the results are in agreement with those determined by the hospital(Kang et al.,2007).It can be seen that the Cu-based modi?ed electrode materials including our obtained Cu2O/TNT electrodes present excellent anti-interference ability against the commonly coexisting species such as uric acid(UA) and ascorbic acid(AA)in real samples.

Most of the electrochemical nonenzymatic glucose sensors based on noble metal are reported to easily lose their activities because of the poisoning by chloride ions,which are abundant in physiological?uids.Thus,the amperometric responses of the Cu2O/TNT electrode to4.0mM glucose in a0.10M NaOH solution and a0.10M NaOH solution containing0.20M NaCl are examined to be89.70μA cmà2and93.5μA cmà2,respectively.The results prove that change in current responses can be enhanced by4.23% in the presence of chloride,implying that the Cu2O/TNT electrode is hardly poisoned by chloride ions.

The reproducibility and stability of the Cu2O/TNT electrode response current were tested.The current density responses (89.32,88.4,95.93,93.55,and92.45μA cmà2)to the addition of 4.0mM of glucose at the same Cu2O/TNT electrode for?ve times minimally changed.The relative standard deviation is 2.89%, indicating that the Cu2O/TNT electrode is stable and can hence be repeatedly used.

The sensor was exposed to air at room temperature.The current density response to the addition of2.0mM glucose at the Cu2O/TNT electrode after26d retained98.61%of its original density of 73.90μA cmà2.The results demonstrated that the electrode can be used in long-term applications.The good reproducibility and long-term stability of Cu2O/TNT electrode can be attributed to both the good stability of Cu2O modi?ed TiO2nanotube arrays in the base solution and to good chemical stability of the materials.

3.4.Discussion for electrocatalytic oxidation of glucose over

Cu2O/TNT electrode

From the CV of the Cu2O/TNT electrode in0.1M NaOH solution with absence of glucose(Fig.3A(c)),a broad and weak oxidation current plateau from about0.30to0.75V is observed,which

can

Fig.3.(A)CVs of bare TNT electrode(a and b)and Cu2O/TNT electrode(c and d)in the absence(a and c)and presence(b and d)of4mM glucose in0.10M NaOH;(B)CVs of the Cu2O/TiO2electrode in0.1M NaOH in different concentrations of glucose:0(a),3.0(b),5.0(c)and8.0mM.Scan rate:50mv/s;(C)CVs of Cu2O/TiO2electrode obtained in0.10M NaOH solution containing2mM glucose at different scan rates(inner to outer):20,50,80,100,120,150,180,and200mV sà1,and the dependence of the oxidation peak current density(at650mV)of glucose on scan rate(left inset).(D)SEM images and CVs of the as-prepared Cu2O/TNTs electrode prepared at different deposition charges:1.0C(a and c)and1.5C(b and d)in the absence and presence of2mM glucose in0.1M NaOH.

M.Long et al./Biosensors and Bioelectronics59(2014)243–250247

be mainly attributed to the formation of different surface species of Cu(I)hydroxide,Cu(II)oxide,or hydroxides.The surface copper hydroxides are easily formed in highly alkaline electrolyte.The appearance of a reversible response at 0.75–0.8V,just prior to the onset of oxygen gas evolution,is assumed to be associated with a Cu(II)/Cu(III)transition,probably in an oxide deposit on the electrode surface.Evidence for the reduction of Cu(III)species has been available for the appearance of an obvious cathodic counterpart at 0.668V in the negative sweep (Fig.3A(c)).We believe that this reduction peak is due to the reduction of Cu(III)species.Similarly,Fleischmann et al.claimed that the copper II/III transition occurs at potentials of 0.65V versus SCE in 0.1M KOH,which is too close to bulk oxygen evolution to be studied in detail.However,the presence of Cu(III)in the apparent charge under the peaks had a clear indication (Wang,et al.,2013a,2013b ).Further-more,the thermodynamic data for copper in an aqueous base are summarized in terms of the following equations:

2CuO t2H tt2e à?Cu 2O tH 2O,E θ

?0.4275V (0.5055)V versus SCE (1)CuO t2H tt2e -?Cu tH 2O,E θ

?0.3285V (0.3675)V versus SCE(2)Cu 2O t2H tt2e -?2Cu tH 2O,E θ?0.2285V versus SCE

(3)

The potential value is quoted in terms of the SCE scale;those in

parentheses refer to cases of the hydrated oxide,CuO ?H 2O or Cu (OH)2.The data outlined above show that the maximum potential reduction of CuO or Cu(OH)2should be no more than 0.5055V,versus SCE.Therefore,the peak at 0.688V cannot be attributed to a Cu(III)/Cu(II)transition.

Upon adding 4.0mM glucose,the broad anodic peak with upward shoulder rapidly increases,which represents glucose irreversible oxidation,as shown in Figs.3A(d)and B.The process involved is clearly irreversible,and no obvious cathodic counterparts in the negative sweep were observed.Moreover,with increasing of glucose concentration,the catalytic currents gradually increase on the Cu 2O/TiO 2electrode (Fig.3B).A decrease in the peak current for the reduction of Cu(III)is found upon increasing glucose concentration.Under these conditions,glucose oxidation may be assumed to be catalyzed by Cu(III)species.The overall behavior observed here is relatively simple,and is somewhat similar to the results on the copper-modi ?ed polyaniline (Farrell and Breslin,2004)and the CuO-modi ?ed TiO 2nanotubes array electrode (Luo et al.,2011)under the same conditions.However,the oxidation currents were higher for Cu 2O/TNT than in Cu/polyaniline and CuO/TNT (Farrell and Breslin,2004;Luo et al.,2011).Based on the above analysis,a schematic illustration of the preparation of Cu 2O/TNT electrode and reaction mechanism of glucose at Cu 2O/TNT electrode is pre-sented in Fig.7

.

Fig.3.(continued )

M.Long et al./Biosensors and Bioelectronics 59(2014)243–250

248

Fig.7describes direct electron transfer between the glucose and the Cu 2O/TNT electrode via oxidation and reduction of Cu 2O/TNT,wherein Cu 2O/TNT acts as a better electron acceptor in the strong alkaline solution.A possible complexation –oxidation pro-cess for glucose oxidation at Cu 2O/TNT electrodes in 0.1M NaOH solute has been proposed on the basis of the cyclic voltammetry data and 1H NMR spectrometry.The broad upward anodic peaks in CVs can be attributed to the adsorption of glucose.Subsequently,the complex intermediate is continuously oxidized on the elec-trode.Finally the glucose is further oxidized at more positive potentials in the strong alkaline solution.

The possible mechanism is presented in Fig.7a –c and described as follows.(a)Glucose in strongly alkaline solutions can be transformed into the enediol compound.This reaction is a slow process that consists of a hydride ion transfer.(b)This enediol structure of glucose may react with cuprous ions to form a complex intermediate,which in turn can be subsequently oxidized at higher potentials (Marioli and Kuwana,1992).The complex of the enediolic structure with surface species of copper was con-tinuously oxidized under the drive of electric ?eld force.The ?rst process involves the oxidation of a complex intermediate.Soluble species were not observed in the CV curves due to the limited effect of the Cu 2O nanoparticle immobilized on the TiO 2nanotube arrays.The second process operates at higher potentials and involves the catalytic oxidation participation of Cu(III),leading to the formation of gluconic acid.(c)Formic acid product is obtained through the cleavage of the gluconic acid C 1–C 2bond in the strong alkaline solution.The continuous oxidation of the glucose complex produces gluconic acid,which further decomposes into formic acid in the 0.1M NaOH solution.The proposed mechanism for the oxidation of glucose at Cu 2O/TNT in alkaline solution is consistent with most experimental ?ndings (Marioli and Kuwana,1992;Luo et al.,2011;Wang et al.,2012;Meng et al.,2013).Difference in reactivity of different electrodes is related to their composition and spatial structure.The Cu 2O/TNT electrodes exhibit numerous attractive features,including high electrocatalytic activity,low detection,good selectivity,stability,and reproducibility.These features are ascribed to three aspects:(a)the large electroactive surface area of Cu 2O modi ?ed helical TiO 2ordered arrays;(b)the excellent synergistic effect due to the combination of Cu 2O nanoparticles and TiO 2nanotube arrays;and (c)the limited effect of the Cu 2O nanoparticle immobilized on the TiO 2nanotube arrays.

4.Conclusions

(a)A non-enzyme glucose sensor was developed using a simple and controllable electrodeposition method.An examination of the morphology suggested that Cu 2O/TNT composites were helical TiO 2nanotube arrays loaded with well-dispersed Cu 2O nanopar-ticles.(b)The synergetic effect of the composite can be attributed to the large speci ?c surface area of Cu 2O loaded on the helical TiO 2nanotubes and the strong electron transfer rate from the Cu 2O/TNT electrode to the glucose.A possible mechanism for glucose continuous oxidation has been proposed on the basis of cyclic voltammetry data and 1H NMR spectrometry.

(c)The nonenzymatic glucose sensing properties of Cu 2O/TNT electrodes exhibited higher sensitivity (14.56μA cm à2mM à1)towards glucose oxidation when compared to electrodes of pure TNT.These properties also exhibited the lowest limit of detection at 62μM,as well as good reproducibility,selectivity,and quick response characteristics.The overall characterization results demon-strate that the Cu 2O/TNT array electrode has high potential for glucose detection.It is found that the catalytic activity of the biosensors depended strongly on the size,morphology,distribution of nanoparticles and the nature of substrate.Therefore,in

the

Fig.4.1H NMR spectra of 30mM glucose solution (a)prior to electrolysis and (b)after a 16h electrolysis period at 0.65V versus SCE at the Cu 2O/TNT

electrode.

Fig.5.Amperometric measurement of Cu 2O/TNT electrode responses to successive addition of 0.5mM glucose in 0.10M NaOH solution at t0.65V.Inset is the calibration curve of the current response versus glucose

concentration.

Fig. 6.Amperometric measurement of the Cu 2O/TNT electrode responses to successive addition of 4mM glucose,0.4mM fructose,0.4mM AA,and 0.4mM UA in 0.10M NaOH solution at t0.65V.

M.Long et al./Biosensors and Bioelectronics 59(2014)243–250249

future work,it is necessary to systematically study the effects of fabrication parameters and morphologies of TiO 2and/or Cu 2O on electrocatalytic oxidation of glucose.Acknowledgments

This work was supported by the National Natural Science Foundation of China (No.51374250)and Hunan Provincial Natural Science Foundation for Innovative Research Groups (No.2013-2).References

Cao,F.,Guo,S.,Ma,H.Y.,Shan,D.C.,Yang,S.X.,Gong,J.A.,2011.Biosens.Bioelectron.

26,2756–2760.

Chen,L.C.,2013.Mater.Sci.Semicon.Process.16,1172–1185.

Dung,N.Q.,Patil,D.,Jung,H.,Kim,D.,2013.Biosens.Bioelectron.42,280–286.El Khatib,K.M.,Hameed,R.M.A.,2011.Biosens.Bioelectron.26,3542–3548.Ensa ?,A.A.,Jafari-Asl,M.,Rezaei,B.,2013.Talanta 103,322–329.Farrell,S.T.,Breslin,C.B.,2004.Electrochim.Acta 49,4497–4503.Guo,S.J.,Wang,E.K.,2011.Nano Today 6,240–264.

Hou,C.T.,Xu,Q.,Yin,L.N.,Hu,X.Y.,2012.Analyst 137,5803–5808.

Huang,L.,Peng,F.,Wang,H.J.,Yu,H.,Geng,W.,Yang,J.,Zhang,S.Q.,Zhao,H.J.,2011.

Mater.Chem.Phys.130,316–322.

Huang,L.,Zhang,S.,Peng,F.,Wang,H.,Yu,H.,Yang,J.,Zhang,S.,Zhao,H.,2010.Scr.

Mater.63,159–161.

Huo,C.,Yang,H.,2012.J.Colloid Interface Sci.384,55–60.

Jiang,L.C.,Zhang,W.D.,2010.Biosens.Bioelectron.25,1402–1407.Kang,.X.,Mai,Z.,Zou,.X.,2007.J.Anal Biochem.363,143–150.

Li,C.L.,Su,Y.,Zhang,S.W.,Lv,X.Y.,Xia,H.L.,Wang,Y.J.,2010.Biosens.Bioelectron.

26,903–907.

Li,X.,Yao,J.,Liu,F.,He,H.,Zhou,M.,Mao,N.,Xiao,P.,Zhang,Y.,2013.Sens.

Actuators B:Chem.181,501–508.

Luo,M.Z.,Baldwin,R.P.,1995.J.Electroanal.Chem.387,87–94.

Luo,S.,Su,F.,Liu,C.B.,Li,J.X.,Liu,R.H.,Xiao,Y.,Li,Y.,Liu,X.N.,Cai,Q.Y.,2011.Talanta

86,157–163.

Luo,Z.J.,Han,T.T.,Qu,L.L.,Wu,X.Y.,2012.Chin.Chem.Lett.23,953–956.Marioli,J.M.,Kuwana,T.,1992.Electrochim.Acta 37,1187–1197.

Meng,F.H.,Shi,W.,Sun,Y.N.,Zhu,X.,Wu,G.S.,Ruan,C.Q.,Liu,X.,Ge,D.T.,2013.

Biosens.Bioelectron.42,141–147.

Park,S.,Park,S.,Jeong,R.A.,Boo,H.,Park,J.,Kim,H.C.,Chung,T.D.,2012.Biosens.

Bioelectron.31,284–291.

Pissinis,D.E.,Sereno,L.E.,Marioli,J.M.,2013.J.Electroanal.Chem.694,23–29.Rahman,M.M.,Ahammad, A.J.S.,Jin,J.H.,Ahn,S.J.,Lee,J.J.,2010.Sensors 10,

4855–4886.

Sun,C.L.,Cheng,W..L,Hsu,T.K.,Chang,C.W.,Chang,J.L.,Zen,J.M.,https://www.360docs.net/doc/fb416563.html,mun.30,91–94.

Wang,C.,Yin,L.,Zhang,L.,Gao,R.,2010a.J.Phys.Chem.C 114,4408–4413.

Wang,G.F.,He,X.P.,Wang,L.L.,Gu,A.X.,Huang,Y.,Fang,B.,Geng,B.Y.,Zhang,X.J.,

2013a.Microchim.Acta 180,161–186.

Wang,G.F.,Wei,Y.,Zhang,W.,Zhang,X.J.,Fang,B.,Wang,L.,2010b.Microchim.Acta

168,87–92.

Wang,L.,Fu,J.Y.,Hou,H.Q.,Song,Y.H.,2012.Int.J.Electrochem.Sci.7,

12587–12600.

Wang,M.Y.,Sun,L.,Lin,Z.Q.,Cai,J.H.,Xie,K.P.,Lin,C.J.,2013b.Energy Environ.Sci.6,

1211–1220.

Xu,F.G.,Deng,M.,Li,G.Y.,Chen,S.H.,Wang,L.,2013.Electrochim.Acta 88,59–65.Yang,J.H.,Zhang,K.T.,Ma,D.,2013.Mater.Lett.97,133–136.

Yang,Z.Y.,Feng,J.S.,Qiao,J.S.,Yan,Y.M.,Yu,Q.Y.,Sun.,KN.,2012.Anal.Methods —UK

4,1924–1926.

Yang,H.M.,Zhang,X.C.,Ouyang,J.,2011.Low-Dimens.Metal Oxide Nanomater.18.Yec,C.C.,Zeng,H.C.,2012.Chem.Mater.24,1917–1929.

Zhang,P.,Zhang,L.,Zhao,G.C.,Feng,F.,2012.Microchim.Acta 176,411–417.

Zhang,S.S.,Zhang,S.Q.,Peng,F.,Zhang,H.M.,Liu,H.W.,Zhao,H.J.,https://www.360docs.net/doc/fb416563.html,mun.13,861–864.

Zhang,X.J.,Wang,G.F.,Zhang,W.,Wei,Y.,Fang,B.,2009.Biosens.Bioelectron.24,

3395–3398.

Zhou,X.M.,Nie,H.,Yao,Z.,Dong,Y.,Yang,Z.,Huang,S.,2012.Sens.Actuators B:

Chem.168,1–7.

Zhu,X.H.,Jiao,Q.F.,Zhang, C.Y.,Zuo,X.X.,Xiao,X.,Liang,Y.,Nan,J.M.,2013.

Microchim.Acta 180,477–483

.

Fig.7.Schematic illustration of the preparation of Cu 2O/TNT electrode and the reaction mechanism of glucose at Cu 2O/TNT electrode.

M.Long et al./Biosensors and Bioelectronics 59(2014)243–250

250