Photoelectrocatalytic Degradation of p-Hydroxybenzoic Acid at the Surface of the Ti/TiO2 Nanotube Array Electrode Using Electrochemical Monitoring

Contents lists available at ScienceDirect

Materials Science in Semiconductor Processing

Materials Science in Semiconductor Processing 31 (2015) 651–657

http://d1369-80

n CorrE-m

journal homepage: www.elsevier.com/locate/mssp

Photoelectrocatalytic degradation of p-hydroxybenzoic acidat the surface of a titanium/titanium dioxide nanotube arrayelectrode using electrochemical monitoring

Reza Ojani n, Akbar Khanmohammadi, Jahan-Bakhsh RaoofElectroanalytical Chemistry Research Laboratory, Department of Analytical Chemistry, University of Mazandaran, Babolsar, Iran

a r t i c l e i n f o

Available online 17 January 2015

Keywords:Photoelectrocatalysisp-Hydroxybenzoic acidHydrodynamic photoamperometryElectrochemical monitoringTi/TiO2-NTA electrode

x.doi.org/10.1016/j.mssp.2014.12.05501/& 2014 Elsevier Ltd. All rights reserved.

esponding author. Tel.: þ98 11 35342301; faail address: [email protected] (R. Ojani).

Downloaded from http://www.elea

a b s t r a c t

In this study, the titanium/titanium dioxide nanotube array (Ti/TiO2-NTA) electrode wasprepared with anodic oxidation of Ti foil electrode. The morphology of Ti/TiO2-NTAelectrode was evaluated with scanning electron microscopy images. The results showedthat the inner diameter of nanotubes is below of 100 nm. The electrochemical behavior ofthe as-prepared Ti/TiO2-NTA electrode was studied using the cyclic voltammetry. Inaddition, a significant photoelectrochemical behavior of the p-hydroxybenzoic acid (p-HBA) was observed on the Ti/TiO2 electrode using the hydrodynamic photoamperometryexperiments. Then, the photoelectrocatalytic (PEC) degradation of the p-HBA wasperformed by this electrode, and compared with photocatalytic (PC), electrooxidation(EC), and direct photolysis by ultra-violet ray. It was found from mechanistic studies thatthe rate constant for the PEC process of Ti/TiO2-NTA electrode was higher than otherdegradation processes. The p-HBA concentration monitoring was carried out with thedifferential pulse voltammetry. Finally, the effects of the solution pH, applied potential,and the p-HBA concentration on the degradation efficiencies were studied and the resultsshowed that the optimum pH for the photoelectrocatalytic degradation was equal to 7.00.The optimum potential and the optimum concentration were about 0.5 V (vs. Ag|AgCl|KCl(3M) as reference electrode) and 0.129 mM in the studied ranges, respectively.

& 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The widespread occurrence of organics in wastewater andrelated environmental hazards has heightened concern overpublic health. The global legislations are putting more empha-sis on recycling of wastewater effluents, which is known to bea strategic approach in a sustainable water management in allparts of the world to minimize the growing water demand ina water-scarce environment [1–3]. In the past few decades,some methods have been used for purification of water anddecreasing of water contaminants. Some of these methodsdeal with biological treatments [4–6]. These methods have

x: þ98 11 35342302.

rnica.ir

complexity due to the use of some bacteria and are compara-tively expensive methods. Other methods that have been usedinclude chemical oxidation with reagents such as ozone,hydrogen peroxide, ozone/UV, hydrogen peroxide/UV andFenton's reagent [7–11]. Nowadays, advanced oxidation pro-cesses (AOP) [12–15], which involve the generation of highlyreactive hydroxyl radical (HO�) have become a promisingwater and wastewater treatment technology for the degrada-tion or mineralization of a wide range of organic contami-nants. Photoactivated processes are a class of AOP, so TiO2-based photocatalytic (PC) and photoelectrocatalytic (PEC)techniques have been proven promising and highly efficientprocesses that can be used for degradation of various recalci-trant organic pollutants, such as dyes, pesticides, herbicides,aromatics, etc., under UV light irradiation [16–20]. The p-hydroxybenzoic acid (p-HBA) is a monohydroxybenzoic acid, a

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657652

phenolic derivative of benzoic acid. The p-HBA is a whitecrystalline solid that is slightly soluble in water. The p-HBAcan be found naturally in Cocos nucifera [21]. It is found inhumans after consumption of green tea [22] and wine [23] asone of the main catechins metabolites. This paper reports forthe first time the results of PEC degradation of p-HBA usingthe Ti/TiO2-NTA electrode as photoanode. This electrode wasprepared through anodization of the Ti foil in the HF solutionfollowed by calcination. To measure the p-HBA concentrationvariation with carbon paste electrode (CPE), the electroche-mical monitoring was performed using differential pulsevoltammetry (DPV). Then, the effective parameters in thePEC degradation such as pH, bias potential, and p-HBAconcentration were examined and optimized.

2. Experimental

2.1. Reagents and materials

The solvent for electrochemical studies was doublydistilled water. The p-hydroxybenzoic acid from Merckwas used as received. Titanium foil (thickness 0.25 mm,assay 99.7%) was purchased from Sigma-Aldrich. Hydro-fluoric acid (99.5%), Ortophosphoric acid (85%), sodiumdihydrogen phosphate (99%), and nitric acid (66%) werepurchased from Fluka. Other materials were from Merck.The buffer solutions contain 0.1 M Na2SO4 as supportingelectrolyte. Buffer solutions were prepared from NaOH andH3PO4 and its salts for pH range of 2.0–10.0.

2.2. Instrumentation

Photoelectrochemical experiments were carried outusing a potentiostat/galvanostat (m-Autolab Type III) and,as shown in Fig. 1, in a single quartz cylindrical photo-reactor cell (3.0 cm diameter�8.0 cm height and 1.8 mmthick). The reactor and the UV lamp were placed in a black

Fig. 1. Schematic diagram of the reactor system: (A) auxiliary electrode,(B) reference electrode, (C) working electrode, (D) magnetic stirrer, (E) UVlamp and (F) galvanostat/potentiostat.

box in order to avoid environment illumination. An Ag|AgCl|KCl (3 M) as reference electrode and a platinumelectrode (Pt) as the auxiliary electrode were used. A4 W medium pressure mercury lamp with maximum UVirradiation was used for excitation of Ti/TiO2 photoelec-trode. The pH of buffers was adjusted with pH-meter(Metrohm, pH 780) to an appropriate pH.

2.3. Fabrication of the Ti/TiO2-NTA electrode

A piece of the Ti foil sheet (3.0 cm�3.0 cm) waspolished with different abrasive papers. Then, the Ti foilwas sonicated in doubly distilled water and chemicallyetched by immersion in HF/HNO3/H2O (1:4:5 V/V/V)mixed solution. Then, the treated Ti sheet was rinsed withacetone and distilled water. The treated Ti sheet served asthe anode and with a Pt electrode, as the cathode, wasplaced in a solution of 0.3% (V/V) hydrofluoric acid and theanodization was performed using a constant 20 V potentialfor 2 h. The freshly Ti/TiO2-NTA electrode was then rinsedwith doubly distilled water and dried in the air. Eventually,it was calcinated in a furnace at 500 1C for 2 h [24].

2.4. Fabrication of carbon paste electrode

A mixture of 1.0 mL of paraffin and 1.0 g of graphitepowder was prepared and blended carefully by handmixing with a mortar and pestle for preparation of carbonpaste. A portion of the homogeneous paste packed into thebottom of a glass tube (internal radius 0.47 cm) and theelectrical contact was provided by a copper wire con-nected to the paste in the inner cavity of the tube. Thesurface of the electrode was smoothed on a white paperand rinsed with doubly distilled water prior to eachexperiment. This electrode was served as the workingelectrode in the electrochemical monitoring of the p-HBAconcentration variation using a DPV.

3. Results and discussion

3.1. Scanning electron microscopy (SEM) of the Ti/TiO2-NTAelectrode

The morphology of the prepared Ti/TiO2-NTA electrodewas assessed by SEM images. The resulted SEM image isshown in Fig. 2. From this figure, we can conclude that thefabricated nanotube arrays are distributed regularly. Theinner diameter of nanotubes is below 100 nm. From theSEM data, it can be concluded that the mean wall thicknessof nanotubes is approximately below 40 nm. The increase inthe electrode potential during the growth of TiO2 nanotubesin fluoride-containing electrolytes usually resulted in anincrease in the nanotube diameter. This has been con-firmed, for example, by SEM images of TiO2 nanotubesgrown in a water–glycerol electrolyte at various appliedpotentials [25]. The functional dependence of the averagenanotube diameter on applied potential is linear across awide range of applied potentials (up to 40 V). The resultsshow the change in the nanotube diameter with increasingpotential. A similar dependence of nanotube diameter onapplied voltage is also true for titanium alloys [26].

Fig. 2. SEM images of Ti/TiO2-NTA electrode.

Scheme 2. Proposed degradation of p-HBA reactions and suggestedproducts.

Scheme 1. Representative schematic for mechanism of PEC degradationat the surface of TiO2.

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657 653

3.2. Photoelectrochemical behavior of p-HBA at the surfaceof Ti/TiO2-NTA electrode

The semiconductors such as TiO2 have conduction bandand valence band and the energy gap between these twobands in TiO2 is about 3.2 eV. Upon UV illumination, theelectrons are excited from the valence band to the conductionband, generating the electron–hole pairs (Scheme 1) [27]

TiO2þhν-e�þhþ (1)

The above reaction produces powerful oxidant (hþ),reacts with other oxidizable species such as target com-pounds, and degrades them. One of the main productscomes from the oxidation of water molecules or hydroxylions, and produces very active hydroxyl radicals

hþþH2O-HO�þHþ (2)

hþþOH̄-HO� (3)

The hydroxide radical is an active oxidant which reactswith substrates (S) and oxidizes that species

SþHO�-oxidation products (4)

On the other hand, the photogenerated electrons reactwith some species in the solution and reduce them toother weaker oxidants such as O2

��(superoxide radical

ion), HO2�, and H2O2 by the following reactions:

e�þO2-O2��

(5)

O2��þHþ-HO2

�(6)

2HO2�-O2þH2O2 (7)

The major limit of photocatalysis efficiency comes fromrecombination of electron–hole pairs or reaction between

the photogenerated electrons and the hydroxyl radicalsthrough the following pathways:

hþþe�-heat (8)

e�þOH�-OH̄

(9)

The applied bias potential to the Ti/TiO2-NTA workingelectrode prevents these two reactions by driving thephotogenerated electrons to the auxiliary electrode,

Fig. 3. Hydrodynamic photoamperometry using methanol as modelcompound in pH¼7.0 and 0.20 V (vs. Ag|AgCl|KCl (3 M)) appliedpotential.

Fig. 4. Cyclic voltammograms of Ti/TiO2-NTA electrode in the pH¼7.0buffer solution containing 0.1 M Na2SO4 as supporting electrolyte at ascan rate of 10 mV s�1: (a) darkness, (b) UV irradiation, (c) darkness andpresence of 0.431 mM p-HBA and (d) UV irradiation and presence of0.431 mM p-HBA.

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657654

resulted to better separation of the electron–hole pairs andthus decreases the electron–hole recombination [28].Oxidation of the p-HBA at the surface of the Ti/TiO2-NTAelectrode and the extent of current was affected by pH,applied potential, and p-HBA concentration. For studyingthese effects, various amounts of pH, bias potential, andthe p-HBA concentration were used. The supposed degra-dation reactions paths are shown in Scheme 2 [29]. Asshown, intermediates may be 3,4-dihydroxybenzoic acid,hydroquinone, and pyrocatechol.

After preparation of the Ti/TiO2-NTA electrode, itsresponse to UV irradiation was studied using hydrodynamicphotoamperometry. This investigation was performed withmethanol as a model compound in studying the stability ofphotocurrent during the time. As shown in Fig. 3, theproduced photocurrent in this investigation had good sta-bility and with switching on the UV lamp, the photocurrentreached to a constant quantity. In order to investigate the p-HBA behavior at the surface of Ti/TiO2-NTA electrode, thephotoelectrochemical activity of this compound was eval-uated using cyclic voltammetry. For this purpose, cyclicvoltammetry experiments were studied in the dark andunder UV light irradiation in the absence and in thepresence of p-HBA at the surface of this electrode (Fig. 4).The tests were conducted in the phosphate buffer solution(pH 7.0). As can be seen in Fig. 4a, in the darkness and in theabsence of p-HBA, the Ti/TiO2 electrode surface producesnegligible photocurrent because of the impossibility of theoxidation of H2O at the surface of the Ti/TiO2-NTA electrode.For oxidation of H2O, an overpotential of 0.78 V(1.23 Vþ0.78 V¼2.01 V at pH¼0) is needed in the darkness.This overpotential value is higher for pH¼7.0, thus the Ti/TiO2-NTA electrode produces negligible photocurrentbecause of the impossibility of the oxidation of H2O [30].However, in the UV irradiation, this electrode produces highanodic photocurrent due to the production of photogener-ated electrons (Fig. 4b). This indicates that photogeneratedelectrons on the Ti/TiO2 electrode could be effectively drivento the auxiliary electrode by applying a positive potentialand produces high current, which would be beneficial toelectron–hole separation. On the other hand, in the presenceof p-HBA and in the darkness, the Ti/TiO2-NTA electrodesurface produces negligible photocurrent (Fig. 4c). Thismeans that the Ti/TiO2-NTA surface had no power tooxidation of the p-HBA in the darkness because the

semiconductor based photoelectrodes need the irradiationto have the degradation power. However, in the UV irradia-tion (Fig. 4d), the conditions are different to the darknesscases, which means that the produced photogeneratedelectron–hole reacts with p-HBA. In this condition, becauseof consumption of photogenerated electrons by p-HBAdegradation or oxidation mechanism (described previouslyin Eqs. (5)–(7)), the photocurrent in outer circuit declined tothe lower amount. This is the main reason why the photo-generated current at the surface of Ti/TiO2-NTA electrode islower in the UV irradiation and in the presence of p-HBA (dcase), compared with the only buffer solution (b case).

3.3. Comparison of various methods for degradation ofp-HBA

To the best of our knowledge, the photoelectrocatalyticdegradation of p-HBA was not studied previously. However,some literatures studied other methods for degradation ofthis compound. For example, Heredia has used the directoxidation by UV irradiation (photolysis), direct oxidation byozone (ozonation), and oxidation by free radicals (mainlyOH.) with 150W UV light source [31]. In these studies, theTiO2 photocatalyst was used as suspension. In these meth-ods, separation and reuse of photocatalyst are difficult, thusin this study, we use photoelectrode that could be usedseveral times and keeps itself efficient. In this study, thedegradation of p-HBA was carried out using the PEC, PC, EC,and direct UV degradation methods. The degradations wereperformed in the pH¼7.0 and E¼0.5 V (vs. Ag|AgCl|KCl(3 M)) bias potential, and initial concentration of p-HBAwas 4.31�10�4 M. The degradations were performedwithin 2 h. For electrochemical monitoring of the p-HBAconcentration variation through degradation stage, every15 min a differential pulse voltammetry (DPV) was recordedusing the carbon paste as the working electrode. The relativep-HBA concentration at any time was determined by Ct/C0where Ct was the p-HBA concentration at any time and C0was the initial concentration of the p-HBA. On the otherhand, there was a direct relation between the p-HBA con-centrations and the peak currents obtained with DPVs, sothat Ct/C0¼ it/i0. The degradation results (Fig. 5) show thatPEC degradation is so effective than other methods, and the

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657 655

p-HBA degradation performance is about 78% in 2 h. TheLangmuir–Hinshelwood kinetic model can describe thephotocatalytic and photoelectrocatalytic reactions. In thisstudy, the experimental results can well fit the first-orderreaction mechanism. If ln(Ct/C0) is drawn vs. time, the slopeof the curvewould be equal to the rate constant (kt), i.e. ln(Ct/C0)¼ f(t)¼kt. The corresponding reaction rate constant ktobtained from the degradation data of p-HBA is listed inTable 1. The experimental results indicate that the reactionrate of p-HBA degradation in the PEC process is quicker thanthat of the PC, direct UV, and EC processes.

Fig. 5. DPVs using carbon paste electrode for monitoring of p-HBA degradation a

3.4. Effective parameters at PEC oxidation of p-HBA

3.4.1. Solution pHpH is one of the factors that have a key effect on the PEC

degradation of organic material through different ways. First,the pH has the primary effect at structure and charge of theorganic compounds. In acidic medium and in the pH lowerthan pKa of the organic compound, the compound canprotonate and gains a positive charge. On the other hand,in the pH higher than pKa the compound is negativelycharged. Moreover, under different pH value conditions, the

t the surface of the Ti/TiO2-NTA electrode. Other conditions are as in Fig. 4.

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657656

hydroxyl groups on the TiO2 surface undergo the followingequilibrium through Lewis acid–base reaction [32]:

pHoPZC: TiOHþHþ⇄TiOH2þ

(10)

pH4PZC: TiOHþOH�⇄TiO�þH2O (11)

The point of zero charge (PZC) of the TiO2 is about pH¼6.2,so at pHoPZC, the TiO2 surface is positively charged, and atpH4PZC the TiO2 surface is negatively charged [33]. It wasreported that the distribution of TiOH is Z80% at3opHo10; TiO�Z20% at pH410 and TiOH2

þZ20% at

pHo3 [34]. On the other hand, for p-HBA with pKa¼4.48, inacidic medium the positively charged form is predominantand in basic medium the negatively charged form is pre-dominant. Thus, the adsorption of negatively charged p-HBAcompound at the surface of the Ti/TiO2-NTA electrode in the3opHo10 is facile, resulting in higher degradation perfor-mance. The obtained DPVs presented in Fig. 6 show that thePEC degradation of the p-HBA at the surface of the Ti/TiO2-NTA electrode in the pH¼7.0 is better than other pHs. Theexperiments were performed in 60 min, and 0.5 V biaspotential was applied.

3.4.2. Effect of applied potentialIt was previously noted that application of bias potential

through the Ti/TiO2-NTA working electrode is one approachin reducing the electron–hole recombination. The positivebias potential causes migration of the photogeneratedelectrons from the Ti/TiO2-NTA working electrode towardauxiliary electrode surface while the photogenerated holesremain at the surface of Ti/TiO2-NTA. The photogeneratedelectrons are trapped by other species such as H2O or O2,thus recombination of electron–hole pairs is decreased. So,

Table 1Rate constant for p-HBA degradation and correlation coefficient forvarious techniques.

Process Rate constant, kt (min�1) Correlation coefficient, R

PEC 0.0117 0.976PC 0.0075 0.974Direct UV 0.0081 0.964EC 0.0008 0.939

Fig. 6. Effect of pH in the PEC degradation of the p-HBA at the surface ofthe Ti/TiO2-NTA electrode at bias potential of 0.5 V (vs. Ag|AgCl|KCl (3 M)).Other conditions are as in Fig. 4.

different values of bias potentials were examined in therange of 0.2–1.0 V vs. reference electrode for 90 min and inthe pH¼7.0. In treatment conditions, results show that, apotential of 0.5 V is optimum for PEC degradation perfor-mance. As shown in Fig. 7, it is obviously seen that the PECdegradation power of this potential is higher than otherpotentials. It is preferred that at 0.5 V potential, the recom-bination of electron–hole pairs reduced effectively byenhanced separation of photogenerated electrons andholes. Therefore, the degradation efficiency increases. How-ever, with the higher bias potential, the effect of degrada-tion decreases. It is because that for a certain thickness ofTi/TiO2-NTA, the number of photogenerated electrons isfinite under a fixed light intensity. Therefore, saturationcurrent is formed when the bias potential is raised to acertain value, then the PEC degradation efficiency is nolonger increased. This phenomenon is similar to thatreported by Zhao [35]. Therefore, the supplied bias potentialhas an optimal value.

3.4.3. Effect of p-HBA concentrationThe p-HBA concentration is one of the main factors in

the PEC degradation. So, different concentrations of the p-HBA were examined to determine the optimum condition.Experiments show that in the lower concentration of p-HBA, the PEC degradation is effective, as can be seen in

Fig. 7. Effect of applied bias potential in the PEC degradation of the p-HBA at the surface of the Ti/TiO2-NTA electrode at optimum pH¼7.0.Other conditions are as in Fig. 4.

Fig. 8. Effect of p-HBA concentration in the PEC degradation of the p-HBAat the surface of the Ti/TiO2-NTA electrode at pH¼7.0 and bias potentialof 0.5 V (vs. Ag|AgCl|KCl (3 M)) in the solution of (a) 0.129 mM, (b)0.216 mM, (c) 0.431 mM and (d) 0.646 mM of the p-HBA. Other condi-tions are as in Fig. 4.

R. Ojani et al. / Materials Science in Semiconductor Processing 31 (2015) 651–657 657

Fig. 8. Because effective sites of photocatalysis surfaceremain unchangeable through the experiments, and thedegradation reactions take place at these sides, withincreased amount of p-HBA, the accumulation of com-pound on the surface of photoelectrode would decline thedegradation rate. This phenomenon could be related tohigh absorption of UV irradiation by p-HBA molecules, inaddition, increased diffraction of UV ray through increasedamounts of the p-HBA molecules in the solution resultedin low amount of photogenerated electron–hole at thesurface of the Ti/TiO2-NTA electrode, and thus the degra-dation performance is declined.

4. Conclusion

This study demonstrates, for the first time, that photo-electrocatalytic degradation is an effective method in degra-dation of p-HBA in the water treatments. On the other hand,the Ti/TiO2-NTA electrode was used successfully for thedegradation of p-HBA with different methods, such as PEC,PC, EC, and direct UV photolysis. It was found that the PECmethod was the best degradation method. In addition,differential pulse voltammetry was found to be able tomonitoring of concentration variation through degradationexperiments. Moreover, it was found that factors such as pH,bias potential, and compound concentration mainly had aneffective role in the PEC degradation. Therefore, variousquantities of above mentioned factors were studied in orderto optimize the operation conditions.

References

[1] J. Radcliff, Desalination 187 (2006) 77–87.[2] V.G. Mitchell, R.G. Mein, T.A. McMahon, Aust. J. Water Resour. 6

(2002) 31–43.[3] A. Fujishima, X. Zhang, D.A. Tryk, Int. J. Hydrog. Energy 32 (2007)

2664–2672.[4] K. Karim, S.K. Gupta, Water Res. 37 (2003) 2953–2959.[5] K. Karim, S.K. Gupta, Bioresour. Technol. 80 (2001) 179–186.[6] Z. She, M. Gao, C. Jin, Y. Chen, J. Yu, Process Biochem. 40 (2005)

3017–3024.

[7] H. Valdés, R.F. Tardón, C.A. Zaror, Water Sci. Technol. 60 (2009)1419–1424.

[8] D. Gates, R. Siegrist, J. Environ. Eng. 121 (1995) 639–644.[9] J. Wu, M.A. Eiteman, S.E. Law, J. Environ. Eng. 124 (1998) 272–277.[10] F.M. Saunders, P.J. Gould, R.C. Southerland, Water Res. 17 (1983)

1407–1419.[11] J. Carriere, J.P. Jones, A.D. Broadbent, Ozone Sci. Eng. 15 (1993)

189–200.[12] S. Parra, S. Malato, J. Blanco, P. Péringer, C. Pulgarin, Water Sci.

Technol. 44 (2001) 219–227.[13] M. Kang, S.Y. Lee, C.H. Chung, S.M. Cho, G.Y. Han, B.W. Kim, K.J. Yoon,

J. Photochem. Photobiol. A: Chem. 144 (2001) 185–191.[14] F.J. Benitez, J.B. Heredia, J.L. Acero, F.J. Rubio, Chemosphere 41 (2000)

1271–1277.[15] C. Galindo, P. Jacques, A. Kalt, J. Photochem. Photobiol. A: Chem. 130

(2000) 35–47.[16] G. Zhanqi, Y. Shaogui, T. Na, S. Cheng, J. Hazard. Mater. 145 (2007)

424–430.[17] Z.Y. Liu, X.T. Zhang, S. Nishimoto, M. Jin, D.A. Tryk, T. Murakami,

A. Fujishima, J. Phys. Chem. C 112 (2008) 253–259.[18] X. Wang, H.M. Zhao, X. Quan, Y.Z. Zhao, S. Chen, J. Hazard. Mater. 166

(2009) 547–552.[19] H.C. Liang, X.Z. Li, J. Hazard. Mater. 162 (2009) 1415–1422.[20] Y. Liu, J. Li, B. Zhou, J. Bai, Q. Zheng, J. Zhang, W. Cai, Environ. Chem.

Lett. 7 (2009) 363–368.[21] G. Dey, M. Chakraborty, A. Mitra, J. Plant Physiol. 162 (2005)

375–381.[22] P.G. Pietta, P. Simonetti, C. Gardana, A. Brusamolino, P. Morazzoni,

A. Bombardelli, Biofactors 8 (1998) 111–118.[23] R.R. Tian, Q.H. Pan, J.C. Zhan, J.M. Li, S.B. Wan, Q.H. Zhang, W.

D. Huang, Molecules 14 (2009) 827–838.[24] F.Y. Oliva, L.B. Avalle, E. Santos, O.R. Camara, J. Photochem. Photobiol.

A: Chem. 146 (2002) 175–188.[25] J.M. Macak, H. Hildebrand, U.M. Jahns, P. Schmuki, J. Electroanal.

Chem. 621 (2008) 254–266.[26] A. Ghicov, S. Aldabergerova, H. Tsuchiya, P. Schmuki, Angew. Chem.

Int. Ed. 45 (2006) 6993–6996.[27] J. Qiu, S. Zhang, H. Zhao, Sens. Actuators B 160 (2011) 875–890.[28] Y. Liu, X. Gan, B. Zhou, B. Xiong, J. Li, C. Dong, J. Bai, W. Cai, J. Hazard.

Mater. 171 (2009) 678–683.[29] R.A. Torres, W. Torres, P. Peringer, C. Pulgarin, Chemosphere 50

(2003) 97–104.[30] A. Valdes, Z.W. Qu, G.J. Kroes, J. Phys. Chem. C 112 (2008) 9872–9879.[31] J.B. Heredia, J. Torregrosa, J.R. Dominguez, J.A. Peres, Chemosphere

42 (2001) 351–359.[32] K. Nohara, H. Hidaka, E. Pelizzetti, N. Serpone, J. Photochem.

Photobiol. A: Chem. 102 (1997) 265–272.[33] J.C. Chou, L.P. Liao, Thin Solid Films 476 (2005) 157–161.[34] M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Water Res. 44 (2010)

2997–3027.[35] X. Zhao, T. Xu, W. Yao, C. Zhang, Y. Zhu, Appl. Catal. B: Environ. 72

(2007) 92–97.