Published: September 19, 2011

r 2011 American Chemical Society 8138 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

ARTICLE

pubs.acs.org/ac

Recyclable and High-Sensitivity Electrochemical Biosensing PlatformComposed of Carbon-Doped TiO2 Nanotube ArraysLiangsheng Hu,† Kaifu Huo,*,†,‡ Rongsheng Chen,† Biao Gao,† Jijiang Fu,† and Paul K. Chu*,‡

†School of Materials and Metallurgy, Wuhan University of Science and Technology, Wuhan 430081, China‡Department of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China

bS Supporting Information

Electrochemical biosensors are extremely useful to the detec-tion and monitoring of different analytes in environmental

and clinical applications such as neurotransmitters, proteins,nucleic acids, and small molecules because of their simplicity,reliability, high selectivity, and sensitivity.1 A highly efficientelectrode with an interface for the charge transfer processconstitutes the core of an electrochemical biosensor and gen-erates potentials and/or currents that are related to the contentsin the test solution.2 Carbon is one of the widely used electro-chemical electrode materials due to its high electrical conductiv-ity, chemical inertness, flexible surface chemistry, and a widepotential window in aqueous media.3,4 In particular, orientedcarbon nanotube and nanofiber arrays are attractive materialsfor highly sensitive and selective electrochemical biosensorsbecause of their high surface-to-volume ratio,5,6 well-definededge plane structure,3 and high degree of biologically accessiblesurface.6,7 However, nanostructured carbon electrodes are proneto surface fouling and passivation by radical intermediates orpolymerization products generated by the electrochemicalreactions between the analytes and electrode surface.8,9

This undesirable phenomenon can lead to significant signalattenuation as well as reduced sensitivity and selectivity overtime.

Several physical and chemical strategies have been developedto refresh the electrode surface in order to improve the reprodu-cibility and stability. For instance, mechanical polishing10,11 hasbeen employed to remove surface contaminants from glassycarbon (GC) and other bulk metal electrodes, but this techniqueis difficult to implement on nanostructured electrodes. Other

processes such as vacuum annealing,12 laser ablation,13 flameetching,9 acid washing,2 chemical oxidative etching,14 and elec-trochemical polarization8 have also been investigated, but theharsh oxidization procedures easily damage and alter the elec-trode microstructure and/or surface chemistry. The backgroundcurrent is affected frequently leading to poor reproducibilityand stability.2 Consequently, it is imperative to design ananostructured electrode having self-cleaning functions ordevelop a self-refreshable technique to prevent electrodefouling.

In a bioelectrochemical analysis, the fouling substances areprimarily organic biomolecules such as serotonin,15 histamine,15

amino acids,16 NADH,17 aniline,18 and so on. These organic bio-molecules can be photocatalytically decomposed into inorganiccompounds of CO2 and H2O on TiO2 and ZnO photocatalystsby ultraviolet (UV) or visible (Vis) irradiation.19,20 This photo-catalytic procedure does not alter the morphology and structureof the photocatalyst surface and enables the use of oxidesemiconducting nanostructures in self-cleaning and refreshableelectrodes. Although one-dimensional CuS,21 CuO,22 ZnO,23

and TiO224�26 nanostructures have been proposed as electrodes

in electroanalysis, to the best of our knowledge, there has been noinvestigation on refreshing electrode surface using photocatalyticapproaches. Highly ordered TiO2 nanotube arrays (TiO2-NTAs)fabricated by anodic oxidation of titanium in a fluoride-containing

Received: June 26, 2011Accepted: September 19, 2011

ABSTRACT: Electrode fouling and passivation are the main reasons forattenuated signals as well as reduced sensitivity and selectivity over time inelectrochemical analysis. We report here a refreshable electrode composed ofcarbon-doped TiO2 nanotube arrays (C-doped TiO2-NTAs), which not only hasexcellent electrochemical activity for simultaneous determination of 5-hydro-xytryptamine and ascorbic acid but also can be easily photocatalytically refreshedto maintain the high selectivity and sensitivity. The C-doped TiO2-NTAs arefabricated by rapid annealing of as-anodized TiO2-NTAs in argon. The residualethylene glycol absorbed on the nanotube wall acts as the carbon source and noforeign carbon precursor is thus needed. The morphology, structure, and composition the C-doped TiO2-NTAs are determined,and the corresponding doping mechanism is investigated by thermal analysis and in situ mass spectroscopy. Because of the highphotocatalytic activity of the C-dopedTiO2-NTAs electrode, the electrode surface can be readily regenerated by ultraviolet or visiblelight irradiation. This photoassisted regenerating technique does not damage the electrode microstructure while rendering highreproducibility and stability.

8139 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

solution27,28 possess large surface areas, good uniformity andconformability, and high porosity. They are potential electrodematerials for electrochemical biosensors because they offer amicroenvironment to increase the quantity and bioactivity ofimmobilized proteins and produce favorable transport pathwaysfor the analytes.24,26,29 However, because of the low conductivity ofTiO2, TiO2-NTAs electrodes exhibit worse electrochemicalactivities29,30 than nanostructured carbon electrodes.

Herein, we present a refreshable electrode composed ofcarbon-doped (C-doped) TiO2-NTAs. It not only exhibitsexcellent electrochemical activity that is better or comparableto that of existing nanostructured C electrodes in the simulta-neous determination of 5-hydroxytryptamine (5-HT) and ascor-bic acid (AA) but also can be readily regenerated photocata-lytically to recover the high selectivity and sensitivity (seeScheme 1). The C-doped TiO2-NTAs are fabricated by quickannealing of the as-anodized TiO2-NTAs in argon in the absenceof foreign carbonaceous precursors. The residual ethylene glycol(EG) absorbed on the nanotube wall during anodization servesas the carbon source, and the C atoms are uniformly distributedalong the entire nanotube forming the C-doped TiO2-NTAs.This self-doping approach is different from the conventionalprocess to produce C-doped TiO2 by heating the anatase TiO2

nanotubes under CO29,31 or acetylene.30,32 In the conventionaltechnique, the C concentration decreases with depth, whereas byusing the simpler and easier process described here, a uniform Cdistribution along the entire nanotube can be accomplished.These highly ordered C-doped TiO2-NTAs have large surfaceareas, well-defined structures, high conductivity, improved elec-tron transfer (ET) efficiency, and enhanced photocatalyticproperties and thus have immense potential in high-sensitivityand selectivity biosensor devices.

’EXPERIMENTAL SECTION

The highly ordered C-doped TiO2-NTAs electrode wasfabricated by rapid annealing of as-anodized TiO2-NTAs inargon. The residual ethylene glycol absorbed on the nanotubewall acts as the carbon source, and no foreign carbon precursor isthus needed. The as-anodized TiO2-NTAs were prepared byelectrochemical anodization of a Ti metal foil in an ammoniumfluoride (NH4F)�ethylene glycol (EG) solution as described inour previous report.33 A graphite foil and Ti foil (2.0 � 1.0 �0.1 cm3, 99.6% purity, Advent Materials) served as the cathodeand anode, respectively. The electrolyte was EG containing0.3 wt % NH4F and 3 vol % doubly distilled water (DDW).Anodization was conducted at 60 V for 30 min using a directcurrent power supply (IT6834, ITECH, Nanjing, China). The

anodized samples were rinsed in DDW, dried in air, and thenannealed at 350�500 �C in a tube furnace in air (designated asthe TiO2-NTAs/air) or Ar (TiO2-NTAs/Ar) at a heating rate of15 �C/min. The morphology, structure, and composition of theTiO2-NTAs/air and TiO2-NTAs/Ar were characterized by field-emission scanning electron microscopy (FE-SEM, FEI Nova 400Nano), X-ray diffraction (XRD, Philips X’Pert Pro), X-rayphotoelectron spectroscopy (XPS, ESCALB MK-II), transmis-sion electron microscopy (TEM, JEOL 2010 equipped with anX-ray energy dispersive spectrometer (EDS)), and Ramanspectroscopy (LabRAM HR). The electrochemical measure-ments were performed in a conventional three-electrode system.TheNTAs were insulated with epoxy resin exposing an area of 81mm2 as the working electrode. An Ag/AgCl electrode served asthe reference electrode, and a Pt foil was the counter electrode.The electrochemical experiments were carried out on a CHI660cpotentiostat (CH Instruments Inc., Shanghai, China). The redoxsystem used in the evaluation of the ET kinetics consisted ofK3[Fe(CN)6] (5 mM) dissolved in 1 M KCl. In the electro-chemical and biosensing study, the phosphate buffer solution(PBS, 0.1 M) was prepared by dissolving NaH2PO4 and Na2H-PO4 in DDW, and the pH value was adjusted to 7.4 by addingH3PO4 and NaOH. 5-HT (99%) and AA (99%) were purchasedfrom Sigma and used without further treatment. Other chemicalswere analytical reagent grade purchased from Beijing ChemicalsCo., Ltd. The fouled electrode surface was regenerated by exposureto UV light for 30 min or visible light for 120 min in DDW at roomtemperature. A high-pressure mercury lamp (1000 W, primarywavelength of 365 nm, 12 mW/cm2) and a 500 W xenon lamp(30 mW/cm2) with a 400 nm cutoff filter were used as the UV andvisible light sources, respectively. In the XPS spectra, the bindingenergies were referenced to the Ar 2p peak at 242.4 eV.

’RESULTS AND DISCUSSION

Figure 1a�c depicts the top-view FE-SEM images of the as-anodized NTAs as well as TiO2-NTAs/air annealed in air and

Scheme 1. Photo-Refreshing Process of the C-Doped TiO2-NTAs Electrode

Figure 1. (a, b, c) FE-SEM images of the top-surface and cross-section(inset) morphology of as-anodized NTAs, TiO2-NTAs/air, and TiO2-NTAs/Ar and (d) spatially resolved EDS elemental maps depicting thedistribution of the constituting elements within the TiO2-NTAs/Ar.

8140 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

TiO2-NTAs/Ar annealed in Ar at 500 �C for 3 h. The nanotubeshave an outer diameter of about 160 nm, and the wall thickness isabout 25 nm. The morphology of NTAs does not changesignificantly after annealing in air or Ar, but the nanotube lengthis reduced slightly by about 0.3 μm compared to the as-anodizedone with a length of 5.4 μm due to partial sintering.29 The mainpeaks in the XRD patterns acquired from TiO2-NTAs/air andTiO2-NTAs/Ar can be indexed to anatase TiO2 except thosefrom the substrate (Figure S-1 in the Supporting Information).The elemental map of the TiO2-NTAs/Ar in Figure 1d showsthat Ti, C, andO are homogeneously distributed in the individualnanotube. The formation of the C-doped TiO2-NTAs is furtherconfirmed by XPS. As shown in Figure 2a, a high C content ofabout 25 at % is observed on the surface and after sputtering for 2min (sputtering rate of approximately 21 nm/min), the Cconcentration is about 5 at % all the way to the nanotube bottom.

This provides evidence that C is uniformly distributed along theentire length of the nanotube. The high-resolution C 1s spectraobtained at different sputtering times are displayed in Figure 2b.In the near surface, the dominant C 1s component at 284.6 eVand shoulder at a lower binding energy of 281.9 eV arecharacteristic of sp2 C�C and C�Ti.34 After sputtering for 4min, the peak positions are unchanged, but the intensity of thepeak at 284.6 eV diminishes dramatically whereas that of the peakat 281.9 eV increases notably. After sputtering for 6min, the peakat 284.6 eV becomes negligible and there is only one peak at281.9 eV. This remains the same until the bottom of the NT. Thepeak at 281.9 eV which is very close to the C 1s peak (281.5 eV)of TiC stems from carbon partially substituting for oxygen inTiO2 resulting in the formation of Ti�Cbonds (TiOxC2�x).

30,34

The high-resolution Ti 2p spectra obtained at different sputter-ing times are depicted in Figure 2c. There are four doublet peakscorresponding to four chemical states of Ti, namely, Ti�C(Ti 2p1/2, 454.9 eV; Ti 2p3/2, 461.0 eV),

34 Ti2+/TiO (Ti 2p1/2, 455.5eV; Ti 2p3/2, 461.7 eV),35 Ti3+/Ti2O3 (Ti 2p1/2, 457.1 eV; Ti2p3/2, 463.0 eV),

35 and Ti4+/TiO2 (Ti 2p1/2, 459.0 eV; Ti 2p3/2,464.7 eV).35 In the near surface, the intensity of the Ti peakscorresponding to TiO2 is very strong, but with sputtering, theintensity of TiO2 decreases but that of Ti�C increases. XPSindicates that the top of the TiO2-NTAs/Ar may be composedof TiO2 together with a small amount of carbonized graphiteand the rest is C-doped TiO2 nanotube. This is corroboratedby Raman spectroscopy (Figure S-2 in the Supporting In-formation). With regard to TiO2-NTAs/Ar, besides the peaksat 394, 513, and 635 cm�1 associated with the anatase phaseTiO2, there are two weak peaks at 1365 and 1598 cm�1

corresponding to the D and G bands of carbon.36 The G band

Figure 2. (a) XPS depth profiles of TiO2-NTAs/Ar. (b) The high-resolution C 1s and (c) Ti 2p spectra acquired after different sputteringtimes, and the sputtering rate is approximately 21.1 nm/min referencedto SiO2.

Figure 3. (a) TG-DTA curves of the preparation process of TiO2-NTAs/Ar and (b) the in situ MS analysis by measurement of the changeof the ion currents for certain atomic mass units during the experimentalperiod.

8141 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

shift to 1598 cm�1 indicates that graphite on the outermostsurface of the nanotube consists of small sp2 carbon clusters.36

This may arise from carbonization of the absorbed EG on theouter surface during annealing in Ar. However, because of thenanoscale space confinement effect, the gas pressure in thenanotube is larger than that on the surface.37 The nanotubularspace provides a favorable nanoscale local environment with highpressure for enhancing the reaction of gaseous carbon speciesfrom the decomposition of EG with TiO2 to form C-dopingTiO2 when the as-anodized NTAs are annealed in argon,resulting in the formation of C-doped TiO2-NTAs. However,during annealing in air, carbon is oxided to form CO2 and so theXPS elemental depth profile acquired from TiO2-NTAs/air

indicates the presence of Ti and O but not C (Figure S-3 inthe Supporting Information).

Figure 3a,b exhibits the thermogravimetric differential thermalanalysis (TG-DTA) results and corresponding in situ massspectra (MS) of the as-anodized TiO2-NTAs annealed in Ar. Aweight loss of about 0.45% resulting from water desorption isobserved from room temperature to about 100 �C. Two clearexothermal peaks emerge at approximately 295 and 360 �C in theDTA curve, accompanied by 1.02% and 4.86% weight losses inthe TG analysis. The weight loss of 1.02% and correspondingsmall exothermal peaks arise from dehydration and the smalldegree of crystallization from anodized hydrated titania toanatase TiO2

27 as well as partial decomposition of EG. Theweight loss of 4.86% and corresponding larger exothermal peakscan be contributed to total EG decomposition and C dopingreactions forming the Ti�C bonds. This is corroborated by theMS results. As shown in Figure 3b, at 295 and 360 �C, the in situmass spectra disclose the presence of H2 (2 amu), CH4 (16 amu),H2O (18 amu), CO (28 amu), and CH3CHO (44 amu). Allthese species are principle products of the decomposition reac-tion of EG.38 The exothermic crystallization reactions on theanodized TiO2-NTAs at 295 �C andC-doping reaction at 360 �Coffer local nanoscale space with high energy and temperature toaid the decomposition of EG and the carbothermal reaction,although the normal decomposing temperature of EG is about500 �C.38 The nanotubular space provides a high pressure forin situ generation of CH4, CO, CH3CHO, and consequently thereaction with TiO2 to form the C-doped TiO2. The TG-DTA-MS results indicate that the annealing temperature of about360 �C is a significant transition point for the C doping reaction,and so the states of carbon in the TiO2-NTAs/Ar samplesannealed at 350 and 400 �C are also evaluated by XPS (FigureS-4 in the Supporting Information). At 350 �C, there is no signalof Ti�C binding while the chemical states of Ti�C can be easilyidentified from the sample annealed at 400 �C. These resultsverify that C-doped TiO2-NTAs can be readily fabricated byannealing as-anodized NTAs prepared in EG in Ar at a tempera-ture higher than 400 �C.

The electrochemical activity of C-doped TiO2-NTAs andTiO2-NTAs associated with the ET reactions between theelectrode and electrolyte containing the Fe(CN)6

3�/4� redoxcouple has been evaluated. Figure 4a shows the cyclic voltam-metry (CV) profiles acquired from the surface of the TiO2-NTAsand C-doped TiO2-NTAs in 1 M KCl with 5 mM K3[Fe(CN)6]at a scanning rate of 100mV/s. The C-doped TiO2-NTAs show apair of well-defined oxidation and reduction peaks at 0.308 and0.240 V, whereas only a reduction peak at�0.235 V indicative ofnegligible reactions is observed from the TiO2-NTAs. The peak-to-peak potential separation (ΔEp) on the C-doped TiO2-NTAselectrode is approximately 68mV. A smallΔEp (close to the idealvalue of 59mV39) indicates a fast ET kinetics for a single-electronelectrochemical reaction at the C-doped TiO2-NTAs electrode.The low ΔEp of 68 mV can be ascribed to the high electro-chemical activity of C-doped TiO2 along with improved electricalconductivity. This is further confirmed by electrochemical im-pedance spectroscopy (EIS) (Figure 4b,c). The charge transferresistance of the Fe(CN)6

3�/4� redox system is near 100 kΩ onthe TiO2-NTAs, while the resistance on the C-doped TiO2-NTAs drops by 4 orders of magnitude to about 30 Ω. We alsoinvestigate the CV curve (Figure S-5 in the Supporting In-formation) acquired from the TiO2-NTAs after coating with alayer of carbon nanotubes. The results reveal no change except a

Figure 4. (a) CV profiles of 5 mM K3[Fe(CN)6] in 1 M KCl solutionobtained at the electrodes of TiO2-NTAs and C-doped TiO2-NTAsusing a scanning rate of 100mV/s. Nyquist plots for TiO2-NTAs (b) andthe C-doped TiO2-NTAs (c) in 1 M KCl containing 5 mM K3[Fe-(CN)6] by applying an ac amplitude of 5 mV under open circuitpotential conditions in a frequency range from 105 to 0.05 Hz.

8142 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

small shift in the reduction peak corresponding to the Fe-(CN)6

3�/4� redox system in comparison with TiO2-NTAs. Itcan thus be concluded that the fast ET kinetics in the Fe-(CN)6

3�/4� redox system mainly stems from the highly con-ductive C-doped TiO2-NTAs. The electrochemical properties ofthe C-doped TiO2-NTAs annealed at different temperatures(350�500 �C) (Figure S-6 in the Supporting Information) showthat a higher temperature leads to a smaller ΔEp value and largerpeak current in the Fe(CN)6

3�/4� redox system. However, if theannealing temperature is higher than 600 �C, the C-doped TiO2-NTAs layer delaminates from the Ti substrate (Figure S-7 in theSupporting Information). Hence, the C-doped TiO2-NTAselectrode annealed at 500 �C at a heating rate of 15 �C/min isselected for subsequent evaluation of the biosening properties.5-HT, also known as serotonin, is an important biogenic aminepresent in many food products and acts as a chemical messengerin the biological system. In conventional electrochemical anal-ysis, 5-HT and AA in the body fluids produce interfering andoverlapping voltammetric responses on metal and unmodifiednanostructured carbon electrodes,40 thus making quantitativeelectrochemical measurements difficult.15 The C-doped TiO2-NTAs produced in this work without any pretreatment ormodification can be directly used as an electrode to distinguish5-HT from AA with high sensitivity and selectivity. Figure 5adepicts the CV response of 0.2 mM 5-HT at the electrodes ofTiO2-NTAs and C-doped C-doped TiO2-NTAs in a 0.1 M pH =7.4 PBS. No oxidation peak is observed from the TiO2-NTAsreflecting the poor electrochemical response to 5-HT. In con-trast, there is an obvious oxidation peak at 0.367 V in C-dopedTiO2-NTAs. Figure 5b displays the CV curves obtained from theC-doped TiO2-NTAs electrode in a 0.1 M pH 7.4 PBS contain-ing 1 mMAA, 0.2 mM 5-HT, and their mixture at a scanning rateof 100 mV/s. Two notable anodic peaks at 0.073 and 0.380 Vcorresponding to direct oxidation of AA and 5-HT, respectively,can be readily discerned. There is a negligible shift in the peakposition compared to the individual response, and the separationbetween the oxidation peaks of AA and 5-HT is about 317 mV.The large separation enables unambiguous determination of5-HT in the presence of a high concentration of AA at theelectrode of C-doped TiO2-NTAs.

The ideal electrochemical biosensing platform should havegood reproducibility and stability in addition to high select-ivity and sensitivity. With regard to 5-HT electrochemicaldetection, the electrode usually suffers from a severe drop inthe amperometric or voltammetric signal over time due to

accumulation of oxidation products of 5-HT passivating theelectrode surface.15,41 This is the main reason for the poorreproducibility and stability as well as reduced sensitivity andselectivity over time for nanostructured C electrodes whendetecting 5-HT. Here, since the C-doped TiO2-NTAs have goodphotocatalytic activity (Figure S-8 in Supporting Information),the accumulated organic species on their surface can be easilydecomposed photocatalytically and renewed by UV or vis lightexposure. Figure 6a shows the CV results of 20 cycles at theoriginal C-doped TiO2-NTAs electrode and the first cycle of thephotocatalytically refreshed C-doped TiO2-NTAs electrode,respectively, in a 0.1 M pH 7.4 PBS containing 0.2 mM 5-HTat a scanning rate of 100 mV/s. It can be observed that theelectrochemical signal vanishes almost completely after 20 cycleson the C-doped TiO2-NTAs, but after UV light irradiation withan intensity of 12 mW/cm2 for 30 min, the sensitivity is restoredwith a relative standard deviation (RSD) of 2.2% (n = 10,Figure 6b). Hence, excellent reproducibility is demonstrated.To confirm the self-renewable capacity of the electrode madeof C-doped TiO2-NTAs, a GC electrode (Figure S-9 in theSupporting Information) is also studied for comparison. Theresults suggest that a fouled GC electrode cannot be refreshedbyUV light exposure. Figure 6c shows the typical amperometricresponse on the C-doped TiO2-NTAs by successive introduc-tion of different concentrations of 5-HT (5�210 μM) into thestirred PBS at 380 mV with and without UV irradiation. Thecurrent response increases with 5-HT concentrations. Thecalibration plot without UV irradiation (black) in Figure 6dillustrates a linear response from 5 to 85 μM with a correlationcoefficient of 0.9938. At higher concentrations of 5-HT, theelectrode is contaminated and the current diminishes. How-ever, under UV irradiation, the calibration plot (red) shows awider linear response from 5 to 150 μM with a correlationcoefficient of 0.9939. The linear range increases significantlyunder UV irradiation, and it can be attributed to photoassistedcleaning of the C-doped TiO2-NTAs electrode surface. Thedetection limit of 5-HT is calculated to be about 4.1 � 10�8 M(S/N = 3), which is lower than or comparable to values in theliterature.15,42 Meanwhile, the enhanced photocurrent mea-sured from the C-doped TiO2-NTAs (red curve in Figure 6c)under UV irradiation further confirms the high photocatalyticactivity.

Parts a and b of Figure 7 are the XPS results obtained from theoriginal C-doped TiO2-NTAs electrode, the fouled electrode suffer-ing from the electrochemical redox reaction of 5-HT, and the

Figure 5. (a) CV profiles of 0.2 mM 5-HT in 0.1 M PBS (pH = 7.4) at TiO2-NTAs and C-doped TiO2-NTAs electrodes. (b) CV curves at C-dopedTiO2-NTAs electrode in PBS with 1 mM AA, 0.1 mM 5-HT, and their mixture. Scan rate: 100 mV/s.

8143 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

renewed electrode after UV photocleaning. In comparison with thatof the original C-doped TiO2-NTAs electrode, the XPS spectrum ofthe fouled electrode shows an obvious peak at 400.3 eV attributableto the N signals in amide or amine (N�C) that originates fromabsorbed and/or electrochemically oxidized products of 5-HT.43

These intermediate species reduce the electrochemical activity oftheC-dopedTiO2-NTAs electrode, and so decreased sensitivity andselectivity are observed over time during the detection 5-HT. Incontrast, after UV light irradiation for 30 min, the N signals in theXPS spectrum disappears and the sensitivity of the electrode after

10 fouling/refreshing runs is still about the same as that in the firstexperiments. DuringUV light irradiation, the absorbed intermediatespecies of 5-HT can be decomposed completely into H2O, CO2,and NOx

19,20,44 without damaging the surface microstructure.Hence, the electrode surface can be refreshed to recover the sensingability. The high photocatalytic activity of the C-doped TiO2-NTAsis also reflected by photodecomposition of rhodamine B (RB) ina solution (Figure S-8 in Supporting Information). These resultsverify that the C-doped TiO2-NTAs electrode is a recyclableplatform for a high-sensitivity biosensor.

Figure 6. (a) CV profiles of 0.2 mM 5-HT in a 0.1 M pH = 7.4 PBS obtained from the original (black) and renewed (red) C-doped TiO2-NTAselectrodes. Scan rate: 100 mV/s. (b) The peak current of different scan number and renew times. (c) Typical current�time response curves at theC-doped TiO2-NTAs electrode through successive addition of different concentrations of 5-HT (5�210 μM) into a stirred 0.1 M PBS (pH = 7.4) at anapplied potential of 380 mV with and without UV irradiation, respectively. The electrodes are not removed between measurements at differentconcentrations without UV irradiation. (d) The calibration curve of the response currents versus concentrations of 5-HT with and without UVirradiation. The error bars represent the response current of three independent experiments.

Figure 7. (a) XPS survey spectra and (b) high-resolution N 1s spectrum acquired from the fresh, fouled, and renewed C-doped TiO2-NTAs electrodes.The XPS spectrum of the fouled electrode shows a large peak at 400.3 eV attributable toN in amide or amine (N�C) that originates from absorbed and/or electrochemically oxidized products of 5-HT. The small peak at about 400.8 eV may come from adsorbed nitrogen on the electrodes.

8144 dx.doi.org/10.1021/ac201639m |Anal. Chem. 2011, 83, 8138–8144

Analytical Chemistry ARTICLE

’CONCLUSIONS

In summary, an electrode composed of C-doped TiO2-NTAsfabricated by rapid annealing of anodized nanotubes exhibitsexcellent electrochemical performance in the simultaneous de-termination of 5-HT and AA that would produce interfering andoverlapping voltammetric responses in conventional electrodes.The residual EG resulting from the anodization process providesthe carbon source, and the nanotube provides the favorablenanoscale space to enhance the carbon-doping reaction. Thehigh photocatalytic activity of the C-doped TiO2-NTAs enablesregeneration of the fouled surface simply by UV or vis lightirradiation without damaging the microstructure, and the surfacecan recover the high selectivity and sensitivity. These outstandingand promising electrochemical properties demonstrate that theC-doped TiO2-NTAs have immense potential in high-sensitivityand high-selectivity biosensor devices.

’ASSOCIATED CONTENT

bS Supporting Information. Additional information asnoted in text. This material is available free of charge via theInternet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: kfhuo@wust.edu.cn (K.H.); paul.chu@cityu.edu.hk(P.K.C.).

’ACKNOWLEDGMENT

This work was jointly supported by National Natural ScienceFoundation of China Grant Numbers 50902104 and 21105077,the Hubei Province Natural Science Foundation Grant No.2010CDB03402, the City University of Hong Kong StrategicResearch Grant (SRG) No. 7008009, and the Hong KongResearch Grants Council (RGC) General Research Funds(GRF) Grant No. CityU 112510.

’REFERENCES

(1) Privett, B. J.; Shin, J. H.; Schoenfisch, M. H. Anal. Chem. 2010,82, 4723–4741.(2) Zoski, C. G.Handbook of Electrochemistry; Elsevier: Oxford, U.K.,

2007.(3) McCreery, R. L. Chem. Rev. 2008, 108, 2646–2687.(4) Yang, W. R.; Ratinac, K. R.; Ringer, S. P.; Thordarson, P.;

Gooding, J. J.; Braet, F. Angew. Chem., Int. Ed. 2010, 49, 2114–2138.(5) Claussen, J. C.; Franklin, A. D.; Haque, A.; Porterfield, D. M.;

Fisher, T. S. ACS Nano 2009, 3, 37–44.(6) Lee, C. S.; Baker, S. E.; Marcus, M. S.; Yang, W. S.; Eriksson,

M. A.; Hamers, R. J. Nano Lett. 2004, 4, 1713–1716.(7) Yu, Z.; McKnight, T. E.; Ericson, M. N.; Melechko, A. V.;

Simpson, M. L.; Morrison, B. Nano Lett. 2007, 7, 2188–2195.(8) Takmakov, P.; Zachek, M. K.; Keithley, R. B.; Walsh, P. L.;

Donley, C.; McCarty, G. S.; Wightman, R. M. Anal. Chem. 2010,82, 2020–2028.(9) Strand, A. M.; Venton, B. J. Anal. Chem. 2008, 80, 3708–3715.(10) Aihara, M.; Komatsu, M. Bull. Chem. Soc. Jpn. 1987, 60,

1911–1912.(11) Thornton, D. C.; Corby, K. T.; Spendel, V. A.; Jordan, J.;

Robbat, A.; Rutstrom, D. J.; Gross, M.; Ritzler, G. Anal. Chem. 1985,57, 150–155.

(12) Fagan, D. T.; Hu, I. F.; Kuwana, T. Anal. Chem. 1985,57, 2759–2763.

(13) Brennan, J. L.; Forster, R. J. J. Phys. Chem. B 2003,107, 9344–9350.

(14) Johnson, D. C.; Lacourse, W. R. Anal. Chem. 1990, 62,589A–597A.

(15) Sarada, B. V.; Rao, T. N.; Tryk, D. A.; Fujishima, A. Anal. Chem.2000, 72, 1632–1638.

(16) Brazill, S. A.; Singhal, P.; Kuhr, W. G. Anal. Chem. 2000,72, 5542–5548.

(17) Wu, L. N.; Zhang, X. J.; Ju, H. X. Anal. Chem. 2007, 79,453–458.

(18) Panizza, M.; Cerisola, G. Chem. Rev. 2009, 109, 6541–6569.(19) Hidaka, H.; Horikoshi, S.; Ajisaka, K.; Zhao, j.; Serpone, N.

J. Photochem. Photobiol., A: Chem. 1997, 180, 197–205.(20) Tran, T. H.; Nosaka, A. Y.; Nosaka, Y. J. Phys. Chem. B 2006,

110, 25525–25531.(21) Lee, H.; Yoon, S. W.; Kim, E. J.; Park, J. Nano Lett. 2007,

7, 778–784.(22) Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2004, 43, 6042–6108.(23) Freeman, R.; Gill, R.; Shweky, I.; Kotler, M.; Banin, U.; Willner,

I. Angew. Chem., Int. Ed. 2009, 48, 309–313.(24) Mun, K. S.; Alvarez, S. D.; Choi, W. Y.; Sailor, M. J. ACS Nano

2010, 4, 2070–2076.(25) Feuz, L.; Jonsson, P.; Jonsson, M. P.; Hook, F. ACS Nano 2010,

4, 2167–2177.(26) Chen, D.; Zhang, H.; Li, X.; Li, J. H. Anal. Chem. 2010, 82,

2253–2261.(27) Grimes, C. A.; Mor, G. K. TiO2 Nanotube Arrays: Synthesis,

Properties, and Applications; Springer: New York, 2009.(28) Ghicov, A.; Schmuki, P. Chem. Commun. 2009, 20, 2791–808.(29) Zhang, Y. H.; Xiao, P.; Zhou, X. Y.; Liu, D. W.; Garcia, B. B.;

Cao, G. Z. J. Mater. Chem. 2009, 19, 948–953.(30) Hahn, R.; Schmidt-Stein, F.; Salonen, J.; Thiemann, S.; Song,

Y. Y.; Kunze, J.; Lehto, V. P.; Schmuki, P. Angew. Chem., Int. Ed. 2009,48, 7236–7239.

(31) Park, J. H.; Kim, S.; Bard, A. J. Nano Lett. 2006, 6, 24–28.(32) Hahn, R.; Ghicov, A.; Salonen, J.; Lehto, V. P.; Schmuki, P.

Nanotechnology 2007, 18, 105604.(33) Zhang, X.M.; Huo, K. F.; Hu, L. S.; Wu, Z.W.; Chu, P. K. J. Am.

Ceram. Soc. 2010, 93, 2771–2778.(34) Gu, D. E.; Lu, Y.; Yang, B. C.; Hu, Y. D. Chem. Commun. 2008,

21, 2453–2455.(35) Chu, C. L.; Hu, T.; Wu, S. L.; Dong, Y. S.; Yin, L. H.; Pu, Y. P.;

Lin, P. H.; Chung, C. Y.; Yeung, K. W. K.; Chu, P. K. Acta Biomater.2007, 3, 795–806.

(36) Ferrari, A. C.; Robertson, J. Phys. Rev. B 2000, 61,14095–14107.

(37) Pan, X. L.; Fan, Z. L.; Chen, W.; Ding, Y. J.; Luo, H. Y.; Bao,X. H. Nat. Mater. 2007, 6, 507–511.

(38) Moldoveanu, S. C. Pyrolysis of Organic Molecules with Applica-tion to Health and Environmental Issues; Elsevier: Oxford, U.K., 2010.

(39) Nugent, J. M.; Santhanam, K. S. V.; Rubio, A.; Ajayan, P. M.Nano Lett. 2001, 1, 87–91.

(40) Michael, A. C.; Borland, L. M. Electrochemical Methods forNeurosscience; CRC Press: New York, 2007.

(41) Anastassiou, C. A.; Patel, B. A.; Arundell, M.; Yeoman, M. S.;Parker, K. H.; O’Hare, D. Anal. Chem. 2006, 78, 6990–98.

(42) Li, Y. X.; Huang, X.; Chen, Y. L.;Wang, L.; Lin, X. Q.Microchim.Acta 2009, 164, 107–112.

(43) Methivier, C.; Lebec, V.; Landoulsi, J.; Pradier, C. M. J. Phys.Chem. C 2011, 115, 4041–4046.

(44) Dolamic, I.; Burgi, T. J. Phys. Chem. C 2011, 115, 2228–2234.

S1

Supporting Information

A Recyclable and High-Sensitivity Electrochemical Biosensing Platform

Composed of Carbon-Doped TiO2 Nanotube Arrays

Liangsheng Hu,† Kaifu Huo,

†,‡* Rongsheng Chen,

† Biao Gao,

† Jijiang Fu,

† and Paul K

Chu‡*

† School of Materials and Metallurgy, Wuhan University of Science and Technology,

Wuhan 430081, China; ‡

Department of Physics and Materials Science, City University of

Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China.

* To whom correspondence should be addressed.

E-mail: kfhuo@wust.edu.cn (K. F. Huo)

E-mail: paul.chu@cityu.edu.hk (Paul K Chu)

S2

Figure S-1. XRD patterns of as-anodized TiO2-NTAs, TiO2-NTAs/air and TiO2-NTAs/Ar.

S3

Figure S-2. Raman scattering spectra acquired from TiO2-NTAs/air and TiO2-NTAs/Ar.

S4

Figure S-3. XPS depth profile of TiO2-NTAs/air (sputtering rate of approximately 21.1

nm/min referenced to SiO2).

S5

Figure S-4. High-resolution C 1s spectra of TiO2-NTAs/Ar: (a) Annealed at 350 oC and (b)

Annealed at 400 oC. Four peaks at 283.3, 284.3, 286.4 and 287.8 eV on the surface and

three peaks at 283.3, 284.3, and 286.4eV in a deeper region can be identified from

TiO2-NTAs/Ar after annealing at 350 °C. The peaks at 283.3, 284.3, 286.4, and 287.8 eV

are attributed to carbide, graphitic carbon, C=O bonds[s1]

, and C-C=O bonds. [s2] For

TiO2-NTAs/Ar annealed at 400 °C, the carbide peak is located at 282.2 eV which can be

attributed to O-Ti-C.

S6

Figure S-5. CV profiles of 5 mM K3[Fe(CN)6] in 1 M KCl solution obtained from the

electrodes made of TiO2-NTAs and TiO2-NTAs/CNTs at a scanning rate of 100 mV/s.

S7

Figure S-6. CV profiles of 5 mM K3[Fe(CN)6] in 1 M KCl solution acquired from the

electrodes made of TiO2-NTAs/Ar with different annealing temperature at a scanning rate of

100 mV/s.

S8

Figure S-7. Optical image of TiO2-NTAs/Ar annealed at 600 oC. If the annealing

temperature is higher than 600 °C, the TiO2-NTAs will delaminate due to the large interface

stress between the nanotube arrays and underlying substrate during rapid heating.

S9

Figure S-8. Concentration versus time for Rhodamine B (RB) under (a) UV and (b) visible

light irradiation in the presence of C-doped TiO2-NTAs and TiO2-NTAs.

The photocatalytic activity of the TiO2-NTAs and C-doped TiO2-NTAs were examined in an

RB aqueous solution as a model organic pollutant. The samples with a size of 1.5 cm2 were

immersed in 20 ml of RB aqueous solution with an initial concentration of 25 µM. The

solution was stirred in the dark for 3 hrs to saturate the photocatalyst with RB and the

photodegradation experiments were performed at the natural pH of the RB dye. The

solution was aspirated continuously with air during the photocatalytic reaction.

S10

Figure S9. CV profiles of 0.2 mM 5-HT in a 0.1 M pH = 7.4 PBS obtained from original

(curve A) and fouled GC electrode. The fouled GC electrode is irradiated for 30 min with

UV light (curve B). The scanning rate is 100 mV/s and the results suggest that the fouled

GC electrode cannot be refreshed by light exposure.

References

[S1] Wang, W.D.; Serp,P.; Kalck, P.; Faria, J. L. J. Mol. Catal. A: Chem. 2005,235, 194-199.

[S2] Adams, J.; Tizazu, G.; Janusz, S.; Brueck, S. R. J.; Lopez, G. P.; Leggett, G. J.

Langmuir, 2010, 26, 13600-13606.