Journal of Electroanalytical Chemistry 765 (2016) 188–196 Contents lists available at ScienceDirect Journal of Electroanalytical Chemistry j ourna l homepage: www.e lsev ie r .com/ locate / je l echem Hydrogen production and simultaneous photoelectrocatalytic pollutant oxidation using a TiO2/WO3 nanostructured photoanode under visible light irradiation T.T. Guaraldo a, V.R. Gonçales b, B.F. Silva a, S.I.C. de Torresi b, M.V.B. Zanoni a,⁎ a Institute of Chemistry, São Paulo State University (UNESP), Rua Prof. Francisco Degni 55, 14800-900 Araraquara, SP, Brazil b Institute of Chemistry, São Paulo University (USP), Av. Prof. Lineu Prestes 748, Cidade Universitária, 05599-970 São Paulo, SP, Brazil ⁎ Corresponding author. E-mail address: boldrinv@iq.unesp.br (M.V.B. Zanoni) http://dx.doi.org/10.1016/j.jelechem.2015.07.034 1572-6657/© 2015 Elsevier B.V. All rights reserved. a b s t r a c t a r t i c l e i n f o Article history: Received 27 March 2015 Received in revised form 19 July 2015 Accepted 21 July 2015 Available online 3 August 2015 Keywords: Bicomponent electrodes TiO2/WO3 templates Photoelectrochemical hydrogen production Dye degradation Photoelectrochemical (PEC) hydrogen production and simultaneous organic waste degradation is a re-emerging field. The main challenge of this technique has been the synthesis of new photoanode materials that are active towards visible light. Coupling close band gap energy oxides can be used to obtain materials with new optical and electronic properties. For this purpose, Ti/TiO2/WO3 electrodes were prepared by electrochemical anodiza- tion followed by templating and cathodic electrodeposition. The nanostructured bicomponent material was used as a photoanode for simultaneous hydrogen generation and organic dye degradation. A good photoactivity response (11 mA cm−2) was obtained under UV and visible light irradiation, when compared to pure TiO2 (8 mA cm−2). Optimization of photoelectrochemical conditions revealed that pH optimization had a major im- pact on H2 production, resulting in satisfactory hydrogen generation efficiency (46%) and dye removal (100% dis- coloration and 85% reduction in TOC). A dye oxidationmechanism is proposed, based on LC-MS/MS analyses. The TiO2/WO3 photoanode could potentially be used for environmental remediation and hydrogen generation under solar irradiation. © 2015 Elsevier B.V. All rights reserved. 1. Introduction The potential expansion of the use of TiO2 as a semiconductor has been held back by its limited absorption in the ultraviolet region. Among themethods reported to produce photoactive TiO2 in the visible light region, a promising strategy involves the coupling of semiconduc- tor materials [1,2]. The coupling of two semiconductors presenting complementary optical properties can enhance electrode performance by improving charge separation (e−/h+), minimizing charge recombi- nation, and increasing photoactivity [3]. The coupling of WO3 and TiO2 can be especially useful for the photoelectrocatalytic oxidation of organ- ic species [4–9]. There are considerable benefits derived from the cou- pling of oxide materials containing close band gap energies, such as TiO2 (3.2 eV) andWO3 (2.8 eV). The valence and conduction band ener- gy diagrams for WO3 and TiO2 bicomponent materials indicate electron injection from the TiO2 conduction band to WO3 is favored, while hole transfer between valence bands occurs in the opposite direction [4]. This can increase the number of holes across the TiO2 surface and en- hance the flow of electrons towards the counter electrode, kept under bias potential, hence improving photoelectrocatalysis efficiency. In ad- dition, bicomponent films have shown greater photochemical activa- tion than individual arrays of WO3 or TiO2 [4]. . Most studies of TiO2/WO3 photoanodes have concerned coatings prepared by electrodeposition and electrosynthesis [8,10], while the preparation of templated TiO2/WO3 nanostructured films has received a reduced amount of attention [7]. Considering recent advances in photoelectrocatalysis is directly related to the morphologies of nano- structured semiconductors, the use of templates could offer a simple and versatile approach for the preparation of dimensionally controlled nanostructured TiO2/WO3 bicomponent films. Advantages of template synthesis include the ability to increase catalyst surface area, improve reaction/interaction between the semiconductor and the electrolyte, consequently enhancing charge transfer efficiency and reducing electron-hole recombination, and extend the absorption spectrum of the catalyst [3]. The cost ofwastewater treatment is amajor concern in industry, and there is a continual search for effective and inexpensive solutions. One attractive option is to employ photoelectrocatalytic techniques in order to explore a useful strategy by recovering the hydrogen generated at the counter electrode. In thesemethods, electrons are driven from the photoanode during the photoelectrocatalytic oxidation of organic com- pounds. Thus, in an ideal system, the organic pollutant is oxidized at the photoanode surface and photogenerated electronsmove through an ex- ternal circuit to the cathode, where reduction of protons occurs under anaerobic conditions. Therefore, the twin environmental benefits of wastematerial elimination and hydrogen generation could be an attrac- tive way of adding value to wastewater treatment. http://crossmark.crossref.org/dialog/?doi=10.1016/j.jelechem.2015.07.034&domain=pdf http://dx.doi.org/10.1016/j.jelechem.2015.07.034 mailto:boldrinv@iq.unesp.br Journal logo http://dx.doi.org/10.1016/j.jelechem.2015.07.034 Unlabelled image http://www.sciencedirect.com/science/journal/ www.elsevier.com/locate/jelechem 189T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 In this work, we report the preparation of Ti/TiO2/WO3 bicompo- nent material by coating nanocavities with WO3 film deposited by cathodic electrodeposition onto organized TiO2 nanotube arrays ob- tained by electrochemical anodization. The aim was to enhance photoelectrocatalytic performance during irradiation from a commer- cial lamp that emitted UV and visible light. The photoanode was employed in the photoelectrocatalytic oxidation of Reactive Black 5 azo dye (RB5), used as an organic pollutant model, with simultaneous hydrogen generation at the Pt cathode. The dye oxidation was moni- tored together with the hydrogen production efficiency. 2. Experimental 2.1. TiO2/WO3 photoanode synthesis The electrode materials were prepared following a sequence of four steps: TiO2 nanotube formation, polystyrene templates self-assembly, WO3 film formation, and templates removal. Titanium foils were sub- mitted to a cleaning and polishing sequence, as described previously [7]. Formation of nanotubes was achieved by electrochemical anodiza- tion (30 V, 50 h) in 0.25 wt.% NH4F in glycerol/water (90:10, v/v) [11]. The electrodes were fired at 450 °C for 30 min in a furnace (Model 650-14 Isotemp Programmable Muffle Furnace, Fisher Scientific). In the next step, the surface was coated with monodispersed polystyrene nanoparticles used as templates (460 nm, Sigma Aldrich), diluted to 0.5% in 1.0 × 10−7 mol L−1 Triton X-100, accord- ing to the procedure described previously [7,12]. Electrodeposition of the WO3 film was performed at −0.45 V bias potential for 45 min [7,13]. The template was chemically removed under magnetic stirring in toluene (Synth) for 24 h. The electrode was then washed in Milli-Q water, dried in an N2 atmosphere, and the WO3 film was fired again at 450 °C for 30 min in a furnace. The TiO2/WO3 bicomponent photoanodewas compared to pure TiO2 electrodes, where TiO2 nanotubes were prepared by anodization according to the procedure described above [11]. 2.2. Characterization of electrodes All the synthesized electrodes were characterized by X-ray dif- fraction (XRD) using a Siemens D5000 X-ray diffractometer with Cu Kα radiation, controlled using Diffrac Plus XRD Commander soft- ware. Morphological examination was performed by field emission gun scanning electron microscopy (FEG-SEM, EDS), using a JEOL 7500F microscope. In order to evaluate the photocurrent response, linear scanning volt- ammograms were recorded in 0.1 mol L−1 sodium sulfate solution in the range from−0.5 to 2.0 V, using a scan rate of 10 mV s−1. The mea- surements were conducted in the absence and presence of irradiation from a commercial 125 W high-pressure mercury lamp (irradiance of 12.55 mW cm−2) with emission in the UV and visible regions [14]. 2.3. Chemicals Reactive Black 5 dye (55% purity) was purchased from Sigma Aldrich. This compound is used commercially as a textile dye. It is clas- sified as harmful to humans according to US and EU legislation and its specifications are CAS 17095-24-8 and C.I. 20505. Dye solutions were prepared by direct dissolution of the powder in ultrapure water. All other reagents were used without previous treatment. 2.4. Photoelectrochemical reactor All the photoelectrochemical measurements were performed in a PEC reactor consisting of two 25 mL compartments separated by a Nafion membrane and fitted with a quartz window (2.5 cm2) (Fig. 1). The PEC reactor was externally irradiated. All the experiments were performed at room temperature (25 °C) using the previously prepared materials as working electrodes, Ag/AgCl as the reference electrode, and a Pt mesh as the counter electrode. The distance between the coun- ter and working electrodes was 4.5 cm. All the photoelectrochemical measurements were carried out using an Autolab PGSTAT 302 potentiostat/galvanostat. 2.5. Hydrogen generation and dye degradation at TiO2/WO3 bicomponent, TiO2, and WO3 electrodes Nitrogen was purged through the PEC reactor for 15 min prior to irradiation, in order to remove oxygen from the solution in each compartment. The amount of H2 produced under irradiation was de- termined using a Thermo TRACEGCUltra gas chromatograph equipped with TCD and FID detectors. A Carboxen-1006 PLOT column was used (30 m × 0.53 mm) and the carrier gas was Ar at a flow rate of 60 mL min−1. A 1 mL syringe was used for sample injection, and the oven, injector, and TCD temperatures were all kept at 150 °C. The anal- ysis time was 8.5 min. Calibration curves were constructed using injec- tion volumes of 20, 40, 60, 80, 100, 200, 300, 400, and 500 μL. The hydrogen evolved was collected in an inverted burette. The photoelectrochemical dye degradation was monitored by means of spectrophotometry (Model 8453, Hewlett-Packard), total or- ganic carbon analysis (TOC-VCPN, Shimadzu, Japan), and high perfor- mance liquid chromatography (LC-MS/MS). Prior to the chromatographic analysis, all the samples were sub- mitted to solid phase extraction, using 3 mL Phenomenex cartridges, in order to remove the electrolyte from the degradation solutions. The extraction procedure used the following sequence: 3 mLmethanol, 3 mL pure water, 500 μL sample, 3 mL pure water, 2 mL ACN/MeOH (50:50), and 1 mL dichloromethane. After extraction, the samples were resuspended in water (150 μL) before injection. The analysis of the dye and its degradation products was performed by full scan enhancedmass spectrometry (EMS), with ion product anal- ysis (EPI) for structural elucidation. Before injection, all samples were diluted in MeOH/H2O (50:50, v/v) containing 0.1% formic acid. The analytes were separated on a Phenomenex Kinetex C-18 column (5 μm, 150 mm × 4.6 mm) coupled to an automatic injector (Model 1200, Agilent) and an HPLC pump (Model 1200, Agilent). Gradient mode was employed for the elution of byproducts, with 0.1% formic acid as a phase modifier in the eluent, over a period of 22 min. The sol- vents used were CH3COONH4 (50 mM in water) (A) and ACN (B), fol- lowing the sequence: 90% A, 10% B (0–2 min); 100% B (2–17 min); 90% A, 10% B (17–22 min). The eluent flow rate was 1.0 mL min−1 and the injection volume was 20 μL. The LC-MS/MS analysiswasperformedusing aModel 1200high per- formance liquid chromatograph (Agilent Technologies) coupled to a 3200 QTRAP quadrupole/linear ion trap mass spectrometer (AB Sciex Instruments) operated in negative ion mode with TurboIon Spray ioni- zation. The spectral data were obtained at a vaporizer temperature of 550 °C, and nitrogen was used as the collision gas. 3. Results and discussion 3.1. Ti/TiO2/WO3 photoanode characterization Fig. 2 shows the complete bicomponent electrode synthesis process and the FEG-SEM micrographs obtained for Ti/TiO2/WO3 photoanodes prepared using the electrochemical deposition and template method [7]. The image of the electrode prepared under optimized conditions (Fig. 2, image 2A) showed the surface completely covered by TiO2 nano- tubes 100 nm in diameter and with thickness of 30–40 μm (measured by profilometry). The final electrodeposited WO3 structure after tem- plate removal showed the presence of nanopores 500 nm in diameter andwith thickness of 40–45 μm(Fig. 2, image 2C). As previously report- ed [7], the XRD spectra (Fig. 2, images 3A and 3C) revealed diffraction Fig. 1. Photoelectrochemical reactor used for dye degradation and hydrogen generation: (1) quartz window, (2) Ti/TiO2/WO3 working electrode, (3) Ag/AgCl reference electrode, (4) Nafion membrane, (5) Pt grid, (6) inverted gas burette, and (7) magnetic stirring. 190 T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 peaks corresponding to TiO2 anatase at 2θ=25, Ti substrate at 2θ=40, and an intermediate monoclinic W18O49 species at 2θ=20.WO3 in the monoclinic phase was identified by the presence of three peaks, at 2θ= 23.06, 23.71, 24.36 aswell as three additional peaks at 33.16, 33.65, and 34.0 from tungsten trioxide phase. Energy dispersive X-ray spectrosco- py (EDX) analysis (Fig. 2, image 3B) confirmed the presence of themain Fig. 2. (1) Ti/TiO2/WO3 template synthesis method, (2) FEG-SEM Adapted from ref. 7). elements composing the Ti/TiO2/WO3 electrode. The EDX spectrum displayed peaks with relative intensities typical of oxygen (E = 0.5 kV), titanium (E = 4.5 kV), and tungsten (E = 1.7 kV). Fig. 3 compares photocurrent vs. potential curves obtained by re- cording linear sweep voltammograms using the Ti/TiO2/WO3 bicom- ponent electrode at 10 mV s−1 in 0.1 mol L−1 Na2SO4, under dark images, (3A and C) XRD analysis, and (3B) EDX analysis. Image of Fig. 1 Image of Fig. 2 Fig. 3. Photocurrent-potential curves obtained in the range from −0.5 to 2.0 V, using a scan rate of 10 mV s−1, for the Ti/TiO2/WO3 bicomponent nanostructured photoanode in the absence of light (A), and for the Ti/TiO2 (B) and Ti/TiO2/WO3 (C) photoanodes irra- diated by a commercial 125 W Hg lamp. The measurements were performed in 0.1 mol L−1 Na2SO4. Fig. 4. Reactive Black 5 dyemolecular structure and absorption spectrum in theUV–Vis re- gion (A), comparative spectra for RB5 degradation using the TiO2 (B) and TiO2/WO3 (C) photoanodes, and the spectrum obtained after 120 min of PEC treatment (D). 191T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 conditions (curve A) and irradiated by UV–Vis light (curve C). For comparison, pure Ti/TiO2 nanotubes voltammogram was also mea- sured, without the WO3 coating (curve B). The TiO2/WO3 bicompo- nent nanostructured photoanode exhibited a higher photocurrent response (11 mA cm−2), compared to the pure TiO2 anode (8 mA cm−2). Both materials were photoactivated by light, because a fraction of the TiO2 nanotubes was not coated by WO3, due to the design of the template model for this bicomponent film. The photo- current curve (curve C) shows contributions of both oxides; WO3 was activated first (from −0.4 to 0.1 V), after which the TiO2 contri- bution (from 0.1 V) can be observed. In the absence of irradiation, the photocurrent response was negligible for all the materials studied. The greater photoactivity of the bicomponent material could be attrib- uted to the contributions of activation by the visible and UV light from the commercial lamp. The emission spectrum of the lamp showed greater intensity in the visible light region, at 435, 550, and 575 nm, compared to the UV region [14]. It was reported previously that TiO2/WO3 templated composites showed enhanced photoactivity, relative to a pure TiO2 photoanode [4,15], and it is known that highly ordered and interconnected nanostructures favor electron transfer along the material [16]. 3.2. Ti/TiO2/WO3 bicomponent electrode applied in degradation of Reactive Black 5 dye Fig. 4 shows typical UV–Vis spectrum recorded for 5.0 × 10−5 mol L−1 RB5 dye in 0.1 mol L−1 Na2SO4 solution at pH 6.0 (curve A). The three main bands observed at 200 (3), 310 (2), and 600 nm (1) could be attributed to aromatic centers (2 and 3) and azo groups present in the molecule structure (1). The intensities of these bands decreased drastically after 20min of photoelectrochemical treat- ment under +1 V potential applied to the Ti/TiO2/WO3 photoanode (curve C). Curve B shows the spectrum obtained for degradation performed with the TiO2 electrode, and curve D shows the absorption spectrumof RB5 after 120min of PEC treatment. Complete discoloration was obtained after both PEC treatments. The color removal constant rate followed pseudo-first order kinetics for both bicomponent (11.8 × 10−2 min−1) and nanotube electrodes (10.6 × 10−2 min−1). TOC removal obtained with the Ti/TiO2/WO3 electrode (85%) was around 12% higher than the TiO2 electrode (73%). The increase in the degradation rate could be attributed to the different charge transfer in bicomponent materials, compared to charge mobility in pure oxides. The photogenerated electrons in the TiO2 conduction band move towards the WO3 conduction band, due to the lower energy level. When holes move from WO3 valence band towards TiO2, they can be- come trapped on the TiO2 surface leading to an increase in the recombi- nation time. The photoelectrocatalytic efficiency of bicomponent material is therefore higher than that of TiO2 [15]. It has been proposed that the charge transfer mechanism in TiO2/WO3 bicomponent elec- trodes involves the trapping of electrons from TiO2 by hexavalent tung- sten during the excitation process, with the tungsten being reduced to the pentavalent state (W5+) [17]. Oxygen then converts the pentava- lent tungsten to the hexavalent state. These steps occur alongside for- mation of the hydroxyl radicals that participate in the degradation process. A schematic representation of the photoelectrocatalytic oxida- tion process at the TiO2/WO3 surface is provided in (Eqs. (1)–(7)) [18]. TiO2WO3 þ hv→WO3 eð Þ þ TiO2 eð Þ ð1Þ TiO2WO3 þ hv→TiO2 hð Þ þ WO3 hð Þ ð2Þ The photogenerated electrons are quickly trapped on the TiO2 surface: TiO2 eð Þ þWO3 eð Þ→TiO2 etð Þ þWO3 etð Þ: ð3Þ Electron pathway: W6þ þ e−↔W5þ−e− ð4Þ Ti4þ þ e−↔Ti3þ−e−: ð5Þ Hole pathway: Ti4þ−OH− þ hþ→Ti4þ−OH:: ð6Þ W6þ−OH− þ hþ →W6þ−OH: ð7Þ The photoelectrochemical behavior of the Ti/TiO2/WO3 bicompo- nent electrode is summarized in Fig. 5. Electrons and holes can be trans- ferred from one to another semiconductor based on their conduction and valence band edge positions (ECB and EVB) [17]. The ECB and EVB values reported for WO3 are +0.25 V and +3.05 V, respectively, while the corresponding values for TiO2 are−0.2V and+3.2V [17]. Re- combination is decreased with increased low band gap energymaterial. Upon irradiation, electrons of the composite material are excited from the valence band of both layers to the conduction band, leaving holes in the valence band. The photogenerated electrons of WO3 are injected Image of Fig. 3 Image of Fig. 4 Fig. 5. Schematic representation of the photoelectrocatalyticmechanism for the TiO2/WO3 photoanode in contact with aqueous electrolyte at pH 7. 192 T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 into the conduction band of TiO2, due to the heterojunction interface ef- fect, easily reaching the substrate surface, and are transferred to the cathode via back contact, producing hydrogen [19,20]. Photogenerated holes, on the other hand, move into theWO3 valence band in the oppo- site direction, driven by the built-in heterojunction field, and are finally carried out by the electrolyte to produce hydroxyl radicals at the photoanode in the photoelectrocatalytic reactor. Fig. 6 compares dye degradation in the absence of applied potential (photocatalysis) and in the absence of catalyst (photolysis). These ex- periments were conducted in order to evaluate the contributions of the material (curve B), the applied potential (curve A), and the photo- lytic treatment with irradiation from the high-pressure mercury lamp (curve C). Dye degradation was monitored using UV–Vis spectropho- tometry and TOC measurements. The results demonstrated the dye color was easily removed in the presence of the bicomponentmaterial (curve B) after 120min of photo- catalytic degradation, suggesting hydroxyl radicals could be responsible for breaking azo bonds, as one of the first steps of the oxidation process. The degradation kinetics was 4.9 ×10−2 min−1 and 48%mineralization was achieved. There was a lower influence of light on RB5 removal (curve C), with the photolytic treatment resulting in only 60% color Fig. 6. RB5 dye degradation performed using different treatment techniques: photoelectrocatalysis (curve A), photocatalysis (curve B), and photolysis (curve C). removal (k= 1.55 × 10−2 min−1) and negligible mineralization. How- ever, photoelectrocatalysis performed at +1.0 V with UV+ Vis irradia- tion (curve A) increased the degradation rate to 11.8 × 10−2 min−1, indicating the large contribution of applied potential to the overall process. 3.3. LC-MS/MS analysis Mass spectrometry analysis was used to evaluate the byproducts formed during the photoelectrocatalytic oxidation of RB5 dye, as well as the degradation pathway. Photoelectrochemical treatment of 7.0× 10−5mol L−1 RB5dye in 0.1mol L−1 Na2SO4 at pH 6 (anodic com- partment) and pH 4 (cathodic compartment) was performed for 150 min at +1.0 V using the Ti/TiO2/WO3 photoanode. The RB5 dye chromatogram presented a peak at a retention time (tr) around 8.0 min, detected asm/z 706 ([M–H–2H2SO4]), as shown in Fig. 9A. The identified ion for RB5 dye was different from the initial molecularmass (991 gmol−1) of the dye structure. This could be attrib- uted to the loss of four Na molecules, resulting in the hydrolyzed form (899.82 g mol−1). The dye structure is protonated in solution ([M– 4Na + 4H+]), with mass of 903 g mol−1, and considering the negative mode ([M–H]−),m/z is 902. The extracted ion chromatogram (XIC) for the RB5 dye is shown in Fig. 9A. The mass spectrum corresponding to the chromatographic band at 8.09 min (Fig. 9B) presented a peak at m/z 706,whose fragment ions spectrum is shown in Fig. 9C, and another signal at m/z 352 (Fig. 9B). The fragment ion spectrum of m/z 706 (Fig. 9C) showed two intense peaks at m/z 524.6 and 447.5, together with other less intense peaks at m/z 642.8, 511.5, 431.4, 344.5, and 278.5. This behavior has not been reported before in the literature, where mass spectrometric analysis has only been described for electro- lytic [21] and biological [22] treatments. The LC-MS/MS results obtained during the photoelectrochemical degradation of RB5 dye were used to track the presence of the dye peak at around 8.0 min. Complete peak area suppression was ob- served, confirming that treatment using TiO2/WO3 caused total remov- al of the dye, as found previously using UV–Vis spectrophotometric analysis. However, intermediate byproducts generated during dye deg- radationwere identified atm/z 137, 183, 200, 210, and 280. Their chem- ical structures are shown in Table 2, together with the MS/MS product ions that confirmed their fragmentation. The concentrations of all the intermediates diminished as the treatment time increased. Three other ions, at m/z 248, 360, and 351, were rapidly formed and then completely degraded. The polar compounds with m/z 200 [22,23] and 183 [24] have already been reported. Based on the above results, it is Fig. 7. Hydrogen volume generated during 120 min of photoelectrocatalytic oxidation of 1.0 × 10−5 mol L−1 RB5 dye under high-pressure mercury lamp irradiation in the pres- ence of Ti/TiO2 (A) and Ti/TiO2/WO3 (B) photoanodes. Inset: Chromatographic signal in- tensity of hydrogen (t = 2.7 min) generated in the cathodic compartment. Image of Fig. 5 Image of Fig. 6 Image of Fig. 7 Fig. 8. Simultaneous hydrogen generation (•) and dye degradation (▪) during photoelectrocatalytic oxidation of RB5 dye. 193T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 clear that RB5 dye was rapidly destroyed, mainly by means of hydroxyl radical oxidation on the bicomponent anode surface. Considering the identified degradation byproducts, a degradation pathway involving parallel reactions is proposed, as shown in Fig. 10. The first step pathway is hydrolysis of the RB5 molecule, leading to m/z 742, followed by a catalytic elimination reaction due to NaOH addition, resulting in m/z 706. Next,m/z 706 undergoes a cyclization Fig. 9. Extracted chromatogram of them/z 706 ion (A), mass spectrum of the chro reaction involving the –NH2 and –OH groups of the hydrolyzed RB5, yielding a [M–2H]2− ion with m/z 361. In the next step of the sequence, hydroxyl radical-mediated desulfonation and hydroxylation generates m/z 436 and 200 ions. m/z 436 ion formation is also mediated by hydroxyl radical attack on the first azo bond, cleaving m/z 706 into m/z 436 and 200 ions, followed by subsequent reactions. Them/z 436 ion undergoes preferen- tial attack of hydroxyl radicals on the second azo group, resulting inm/z 268. On the other hand, them/z 200 ion loses one hydroxyl group, lead- ing tom/z 183. The formation of alkylsulfonyl phenolic compounds has been reported previously by Méndez-Martínez and collaborators [25]. The presence of byproducts detected even after 150 min of treatment indicates that recalcitrant material was obtained after 120 min of treatment. 3.4. Hydrogen generation Experiments were carried out to measure the simultaneous gen- eration of hydrogen at the cathode (Pt grid, area = 0.5 cm2). Fig. 7 il- lustrates the amount of hydrogen produced at the cathode during 120 min of photoeletrocatalytic oxidation of 1.0 × 10−5 mol L−1 RB5 dye, under E = 1.0 V and UV + Vis irradiation of the Ti/TiO2/ WO3 photoanode. The generation of hydrogen was confirmed by GC-TCD analysis, as shown in Fig. 7 inset. For comparison purposes, the results obtained using Ti/TiO2 as the photoanode are also shown. Hydrogen production in the cathodic compartment increased as a function of photoelectrolysis time, for all the photoanodes tested, but matographic band at 8.09 min (B), and m/z 706 fragment ion spectrum (C). Image of Fig. 8 Image of Fig. 9 Table 1 Dye discoloration, oxidation kinetics, TOC removal, volume of hydrogen, number of moles of hydrogen produced, and estimated hydrogen conversion efficiency for the degra- dation of 5.0 × 10−5 mol L−1 RB5 dye in 0.1 mol L−1 Na2SO4 at 1 V potential using the bicomponent photoanode. pHanode pHcathode % discoloration k/min−1 % TOC removal VH2 (mL) Moles of H2 produced H2 estimated efficiency 2 4 100% 38.1 × 10−2 95% 0.50 36.0 × 10−5 15% 4 4 95% 12.5 × 10−2 b50% 0.50 11.1 × 10−5 30% 6 4 100% 11.8 × 10−2 85% 0.65 5.0 × 10−5 46% 8 4 85% 11.1 × 10−2 b50% 1.00 4.2 × 10−5 43% 10 4 90% 15.9 × 10−2 b50% 0.50 5.6 × 10−5 30% 194 T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 was higherwhen the dyewas oxidized at the bicomponent anode, com- pared to the pure TiO2 anode. This indicated that there was an improve- ment in electron flow to the cathode when the photoelectrocatalysis of RB5 dye was performed using Ti/TiO2/WO3, confirming that the bicom- ponent anode amplified charge generation/separation. In order to understand the formation of hydrogen in the coupled process, the photoelectrocatalytic oxidation of RB5 dye and hydrogen formation was carried out varying the pH in the cathode and anode compartments, the dye concentration, and the applied potential. Alteration of the potential applied to the photoanode from 1.0 to 1.3 V decreased the efficiency of hydrogen production at the cathode by around 50%, although slightly higher RB5 dye discoloration was ob- tained at 1.3 V in the presence of the Ti/TiO2/WO3 photoanode. This be- havior could be explained by the occurrence of water splitting at 1.23 V, which can be competitive with the water oxidation process on the photoanode [26]. At the same time, the increase in potential from 1.0 to 1.3 V led to a significant decrease in hydrogen generation efficiency, due to a decrease in the photoconversion efficiency of the Ti/TiO2/ WO3 photoanode [15]. The influence of the pH in the anodic compartment, in the range of pH 2.0–10.0, was investigated for photoelectrochemical oxidation of Table 2 Byproducts identified during RB5 photoelectrocatalysis: proposed fragment ion struc- tures, retention times, m/z ratios, and MS/MS product ions. Proposed structure (tR)/min [M–H]− or m/z Productions MS/MS 9.4 436 247.4 182.1 7.9 361 247.2 216.0 182.1 164.0 80.4 Not identified 7.98 360 261.6 80.0 6.4 248 233.9 221.2 150.3 80.0 Not identified 6.9 210 134.9 119.1 10.5 200 184.2 170.1 156.1 7.2 183 156.1 107.9 92.0 RB5 dye and concomitant hydrogen generation. The pH was kept con- stant during the degradations by correction with diluted solutions of H2SO4 and NaOH. In the cathodic compartment, the pHwas initially ad- justed to 4.0, and a pH of 2.0 was also tested. Table 1 shows the results obtained for dye discoloration, oxidation kinetics, and TOC removal dur- ing 120 min of photoelectrocatalysis at the Ti/TiO2/WO3 photoanode operated at 1 V, using 5.0 × 10−5 mol L−1 of RB5 dye in 0.1 mol L−1 Na2SO4. Also shown in Table 1 are the numbers of moles of hydrogen formed and the estimated hydrogen conversion efficiencies at the pH values used. Dye discoloration was observed at all pH values tested, but higher discolorationwas obtainedwith pH 2 at the anode and pH 4 at the cath- ode. However, the minimum hydrogen generation efficiency was ob- tained under these conditions. As reported previously [27], hydrogen generation is favored when there is a pH gradient between the anodic and cathodic compartments. When the same acid pH is used in both compartments, a small amount of H2 is produced. This is known as a chemically biased system [27]. If a pH difference is not maintained, both dye degradation and hydrogen evolution are impaired. Hydrogen evolution and TOC removal were lower at acid pH values of 2 and 4 (anodic compartment). On the other hand, dye degradation was higher under these conditions. Similarmineralizationwas observed for pH values of 6, 8, and 10, although dye discoloration was lower than under acidic conditions, while hydrogen generation was increased. Re- duction of the cathodic pH to 2.0 did not significantly improve hydrogen generation. Under alkaline conditions, hydroxylated byproducts can be formed by reaction between dye degradation products and hydroxyl radicals present in the medium. These byproducts are more difficult to oxidize. Under acidic conditions, aliphatic short chain hydrocarbons are likely to be formed, which can be oxidized more easily, hence favoring discoloration. The results showed that the best experimental conditions for dye degradation and hydrogen evolution were pH 6 in the anodic compart- ment and pH 4 in the cathodic compartment. The nanostructured TiO2/ WO3 bicomponent electrodes demonstrated efficient hydrogen genera- tion, at a rate of 0.33mL h−1 cm−2, corresponding to an estimated effi- ciency of around 46%. 3.5. Simultaneous dye degradation and hydrogen generation Fig. 8 illustrates the performance of the system for simultaneous dye degradation and hydrogen generation during photoelectrocatalytic ox- idation conducted with 1.0 × 10−5 mol L−1 of RB5 dye, using the Ti/ TiO2/WO3 photoanode operated at 1 V bias potential under UV + Vis light. Complete dye degradation was obtained after 120 min of treat- ment, with 100% discoloration and around 85% TOC removal. There was concomitant formation of hydrogen, with a maximum of 5.0 × 10−5 mol cm−2 of hydrogen generated after 120 min of treat- ment, corresponding to 46% conversion efficiency. Table 3 shows a comparison of the charge consumed and the estimated hydrogen generation efficiency when the PEC treatment was performed with the TiO2 anode. Lower charge consumption was observed for the Ti/TiO2/WO3 material, indicating that it could be an efficient means Unlabelled image Unlabelled image Fig. 10. Proposed degradation pathways for photoelectrocatalytic oxidation of RB5 dye. 195T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 of generating hydrogen as a renewable energy resource, while at the same time eliminating undesirable organic material. 4. Conclusions A templating method was successfully used to synthesize a bicom- ponent electrode based on Ti/TiO2/WO3, where TiO2 nanotubes were used as a platform for WO3 deposition. The electrode could be photoactivated by visible and ultraviolet irradiation, and provided an enhanced photocurrent response, relative to plain Ti/TiO2 electrode. The new material enabled more effective photoelectrocatalytic oxida- tion of Reactive Black 5 dye (RB5) under UV–Vis irradiation with an ap- plied potential of +1 V, resulting in 100% discoloration and 85% TOC removal after 120 min of treatment. LC-MS/MS analysis indicated the oxidation involved progressive degradation of the dye and formation of two main compounds as final products, with m/z of 268 and 183. Monitoring of simultaneous hydrogen generation due to electron flow to the cathode showed production of 5.0 × 10−5 mol cm−2 of the gas, using a two compartment reactor operated using 0.1 mol L−1 Na2SO4, anode pH 6.0, cathode pH 4.0, +1.0 V bias potential, and UV + Vis irradiation. The findings demonstrated that photoelectrocatalysis using the TiO2/WO3 bicomponent material as the photoanode provided faster dye degradation, greater mineralization, and improved hydrogen generation, compared to pure oxide electrode. Most of the byproducts formed from RB5 dye were consumed during the PEC Image of Fig. 10 Table 3 Measured hydrogen volume, charge consumed, and estimated hydrogen generation efficiency after 120 min of photoelectrocatalysis of 1.0 × 10−5 mol L−1 RB5 dye in 0.1 mol L−1 Na2SO4 at +1 V potential for each photoanode studied. Electrode VH2/mL Consumed charge/C H2 mols produced Estimated H2 efficiency (%) TiO2 0.60 10.80 5.6 × 10−5 43 TiO2/WO3 0.65 9.84 5.0 × 10−5 46 196 T.T. Guaraldo et al. / Journal of Electroanalytical Chemistry 765 (2016) 188–196 treatment. Therefore, the proposed photoelectrochemical method using a bicomponent TiO2/WO3 electrode could be a promising alter- native for both environmental remediation and energy generation under solar irradiation. Acknowledgements The authors thank FAPESP (process no. 2009/17346-1, 2008/10449- 7, 2008/10449-7), CAPES, and CNPq (446245/2014-3) for financial sup- port of this work. We are indebted to Dra. Sandra I. Maintinguer and Renan Pachiega of IQAr-UNESP for the GC-TCD analyses. References [1] M. Pelaez, N.T. Nolan, S.C. Pillai, et al., A review on the visible light active titanium dioxide photocatalysts for environmental applications, Appl. Catal. B Environ. 125 (2012) 331–349, http://dx.doi.org/10.1016/j.apcatb.2012.05.036. [2] Y. Zhang, X. Xiong, Y. Han, et al., Photoelectrocatalytic degradation of recalcitrant or- ganic pollutants using TiO2 film electrodes: an overview, Chemosphere 88 (2) (2012) 145–154, http://dx.doi.org/10.1016/j.chemosphere.2012.03.020. [3] G.G. Bessegato, T.T. Guaraldo, M.V.B. Zanoni, Enhancement of photoelectrocatalysis efficiency by using nanostructured electrodes, Modern Electrochemical Methods in Nano, Surface and Corrosion ScienceIntech 2014, pp. 271–319, http://dx.doi. org/10.5772/58333. [4] J. Georgieva, E. Valova, S. Armyanov, N. Philippidis, I. Poulios, S. Sotiropoulos, Bi- component semiconductor oxide photoanodes for the photoelectrocatalytic oxida- tion of organic solutes and vapours: a short review with emphasis to TiO2–WO3 photoanodes, J. Hazard. Mater. 211–212 (2012) 30–46, http://dx.doi.org/10.1016/j. jhazmat.2011.11.069. [5] J. Georgieva, S. Sotiropoulos, S. Armyanov, N. Philippidis, I. Poulios, Photoelectrocatalytic activity of bi-layer TiO2/WO3 coatings for the degradation of 4-chlorophenol: effect of morphology and catalyst loading, J. Appl. Electrochem. 41 (2) (2011) 173–181, http://dx.doi.org/10.1007/s10800-010-0221-8. [6] E. Valova, J. Georgieva, S. Armyanov, et al., Morphology, structure and photo- electrocatalytic activity of TiO2/WO3 coatings obtained by pulsed electrodeposition onto stainless steel, J. Electrochem. Soc. 157 (5) (2010) D309–D315, http://dx.doi. org/10.1149/1.3356001. [7] T.T. Guaraldo, T.B. Zanoni, S.I.C. de Torresi, et al., On the application of nanostruc- tured electrodes prepared by Ti/TiO2/WO3 “template”: a case study of removing tox- icity of indigo using visible irradiation, Chemosphere 91 (5) (2013) 586–593, http:// dx.doi.org/10.1016/j.chemosphere.2012.12.027. [8] J. Georgieva, S. Armyanov, E. Valova, I. Poulios, S. Sotiropoulos, Enhanced photocat- alytic activity of electrosynthesised tungsten trioxide–titanium dioxide bi-layer coatings under ultraviolet and visible light illumination, Electrochem. Commun. 9 (3) (2007) 365–370, http://dx.doi.org/10.1016/j.elecom.2006.09.028. [9] J. Georgieva, S. Armyanov, E. Valova, N. Philippidis, I. Poulios, S. Sotiropoulos, Photoelectrocatalytic activity of electrosynthesised tungsten trioxide–titanium dioxide bi-layer coatings for the photooxidation of organics, J. Adv. Oxid. Technol. 11 (2) (2008) 300–307. [10] N. De Tacconi, C. Chenthamarakshan, K. Rajeshwar, T. Pauporte, D. Lincot, Pulsed electrodeposition of WO3–TiO2 composite films, Electrochem. Commun. 5 (3) (2003) 220–224, http://dx.doi.org/10.1016/S1388-2481(03)00021-3. [11] J.C. Cardoso, T.M. Lizier, M.V.B. Zanoni, Highly ordered TiO2 nanotube arrays and photoelectrocatalytic oxidation of aromatic amine, Appl. Catal. B Environ. 99 (1–2) (2010) 96–102, http://dx.doi.org/10.1016/j.apcatb.2010.06.005. [12] V.R. Gonçales, M.P. Massafera, T.M. Benedetti, D.G. Moore, S.I. Córdoba de Torresi, R.M. Torresi, Nanostructured thin films obtained by electrodeposition over a colloi- dal crystal template: applications in electrochemical devices, J. Braz. Chem. Soc. 20 (4) (2009) 663–673, http://dx.doi.org/10.1590/S0103-50532009000400010. [13] J.R.C. Rocha, T.L. Ferreira, R.M. Torresi, M. Bertotti, An analytical application of the electrocatalysis of the iodate reduction at tungsten oxide films, Talanta 69 (1) (2006) 148–153, http://dx.doi.org/10.1016/j.talanta.2005.09.010. [14] G.G. Bessegato, J.C. Cardoso, B.F. Da Silva, M.V.B. Zanoni, Enhanced photoabsorption properties of composites of Ti/TiO2 nanotubes decorated by Sb2S3 and improve- ment of degradation of hair dye, J. Photochem. Photobiol. A Chem. 276 (2013) 96–103, http://dx.doi.org/10.1016/j.jphotochem.2013.12.001. [15] K. Liu, Y.-C. Hsueh, C.-Y. Su, T. Perng, Photoelectrochemical application of mesopo- rous TiO2/WO3 nanohoneycomb prepared by sol–gel method, Int. J. Hydrog. Energy 38 (19) (2013) 7750–7755, http://dx.doi.org/10.1016/j.ijhydene.2013.04.046. [16] E. Yang, J. Shi, H. Liang, W. Cheuk, Coaxial WO3/TiO2 nanotubes/nanorods with high visible light activity for the photodegradation of 2,3-dichlorophenol, Chem. Eng. J. 174 (2–3) (2011) 539–545, http://dx.doi.org/10.1016/j.cej.2011.09.030. [17] S.A. Singh, G.Madras, Photocatalytic degradation with combustion synthesizedWO3 and WO3TiO2 mixed oxides under UV and visible light, Sep. Purif. Technol. 105 (2013) 79–89, http://dx.doi.org/10.1016/j.seppur.2012.12.010. [18] L. Yang, Y. Xiao, S. Liu, et al., Photocatalytic reduction of Cr(VI) on WO3 doped long TiO2 nanotube arrays in the presence of citric acid, Appl. Catal. B Environ. 94 (1–2) (2010) 142–149, http://dx.doi.org/10.1016/j.apcatb.2009.11.002. [19] S. Choudhary, S. Upadhyay, P. Kumar, et al., Nanostructured bilayered thin films in photoelectrochemical water splitting — a review, Int. J. Hydrog. Energy 37 (24) (2012) 18713–18730, http://dx.doi.org/10.1016/j.ijhydene.2012.10.028. [20] H. Zhang, G. Chen, D.W. Bahnemann, Photoelectrocatalytic materials for environ- mental applications, J. Mater. Chem. 19 (2009) 5089–5121, http://dx.doi.org/10. 1039/b821991e. [21] A.J. Méndez-Martínez, M.M. Dávila-Jiménez, O. Ornelas-Dávila, et al., Electrochemi- cal reduction and oxidation pathways for Reactive Black 5 dye using nickel elec- trodes in divided and undivided cells, Electrochim. Acta 59 (2012) 140–149, http://dx.doi.org/10.1016/j.electacta.2011.10.047. [22] X. Wang, X. Cheng, D. Sun, Interaction in anaerobic biodecolorization of mixed azo dyes of Acid Red 1 and Reactive Black 5 under batch and continuous conditions, Col- loids Surf. A Physicochem. Eng. Asp. 379 (1–3) (2011) 127–135, http://dx.doi.org/ 10.1016/j.colsurfa.2010.11.065. [23] X. Wang, X. Cheng, D. Sun, H. Qi, Biodecolorization and partial mineralization of Re- active Black 5 by a strain of Rhodopseudomonas palustris, J. Environ. Sci. 20 (10) (2008) 1218–1225, http://dx.doi.org/10.1016/S1001-0742(08)62212-3. [24] A. Plum, A. Rehorek, Strategies for continuous on-line high performance liquid chromatography coupled with diode array detection and electrospray tandem mass spectrometry for process monitoring of sulphonated azo dyes and their in- termediates in anaerobic–aerobic bioreactors, J. Chromatogr. A 1084 (1–2) (2005) 119–133, http://dx.doi.org/10.1016/j.chroma.2005.03.001. [25] A.J. Méndez-Martínez, M.M. Dávila-Jiménez, O. Ornelas-Dávila, et al., Electrochemi- cal reduction and oxidation pathways for Reactive Black 5 dye using nickel elec- trodes in divided and undivided cells, Electrochim. Acta 59 (2012) 140–149, http://dx.doi.org/10.1016/j.electacta.2011.10.047. [26] Y. Xin, M. Gao, Y. Wang, D. Ma, Photoelectrocatalytic degradation of 4-nonylphenol in water with WO3/TiO2 nanotube array photoelectrodes, Chem. Eng. J. 242 (2014) 162–169, http://dx.doi.org/10.1016/j.cej.2013.12.068. [27] L.J. Minggu, W.R. Wan Daud, M.B. Kassim, An overview of photocells and photoreactors for photoelectrochemical water splitting, Int. J. Hydrog. Energy 35 (11) (2010) 5233–5244, http://dx.doi.org/10.1016/j.ijhydene.2010.02.133. http://dx.doi.org/10.1016/j.apcatb.2012.05.036 http://dx.doi.org/10.1016/j.chemosphere.2012.03.020 http://dx.doi.org/10.5772/58333 http://dx.doi.org/10.1016/j.jhazmat.2011.11.069 http://dx.doi.org/10.1016/j.jhazmat.2011.11.069 http://dx.doi.org/10.1007/s10800-010-0221-8 http://dx.doi.org/10.1149/1.3356001 http://dx.doi.org/10.1016/j.chemosphere.2012.12.027 http://dx.doi.org/10.1016/j.elecom.2006.09.028 http://refhub.elsevier.com/S1572-6657(15)30037-0/rf0045 http://refhub.elsevier.com/S1572-6657(15)30037-0/rf0045 http://refhub.elsevier.com/S1572-6657(15)30037-0/rf0045 http://refhub.elsevier.com/S1572-6657(15)30037-0/rf0045 http://dx.doi.org/10.1016/S1388-2481(03)00021-3 http://dx.doi.org/10.1016/j.apcatb.2010.06.005 http://dx.doi.org/10.1590/S0103-50532009000400010 http://dx.doi.org/10.1016/j.talanta.2005.09.010 http://dx.doi.org/10.1016/j.jphotochem.2013.12.001 http://dx.doi.org/10.1016/j.ijhydene.2013.04.046 http://dx.doi.org/10.1016/j.cej.2011.09.030 http://dx.doi.org/10.1016/j.seppur.2012.12.010 http://dx.doi.org/10.1016/j.apcatb.2009.11.002 http://dx.doi.org/10.1016/j.ijhydene.2012.10.028 http://dx.doi.org/10.1039/b821991e http://dx.doi.org/10.1039/b821991e http://dx.doi.org/10.1016/j.electacta.2011.10.047 http://dx.doi.org/10.1016/j.colsurfa.2010.11.065 http://dx.doi.org/10.1016/S1001-0742(08)62212-3 http://dx.doi.org/10.1016/j.chroma.2005.03.001 http://dx.doi.org/10.1016/j.electacta.2011.10.047 http://dx.doi.org/10.1016/j.cej.2013.12.068 http://dx.doi.org/10.1016/j.ijhydene.2010.02.133 Hydrogen production and simultaneous photoelectrocatalytic pollutant oxidation using a TiO2/WO3 nanostructured photoanode under visible light irradiation 1. Introduction 2. Experimental 2.1. TiO2/WO3 photoanode synthesis 2.2. Characterization of electrodes 2.3. Chemicals 2.4. Photoelectrochemical reactor 2.5. Hydrogen generation and dye degradation at TiO2/WO3 bicomponent, TiO2, and WO3 electrodes 3. Results and discussion 3.1. Ti/TiO2/WO3 photoanode characterization 3.2. Ti/TiO2/WO3 bicomponent electrode applied in degradation of Reactive Black 5 dye 3.3. LC-MS/MS analysis 3.4. Hydrogen generation 3.5. Simultaneous dye degradation and hydrogen generation 4. Conclusions Acknowledgements References