Synthesis of novel Zn_0.5Mg_0.5Fe₂O₄@TiO₂ nanotube arrays with enhanced photoelectrocatalytic properties

Abstract

A novel nanocomposite electrode with tight and vertically aligned Zn_0.5Mg_0.5Fe₂O₄@TiO₂ nanotube arrays (NTs) has been successfully synthesized via an ultrasonically assisted electrodeposition strategy. The scanning electron microscopy (SEM), transmission electron microscopy (TEM) and X-ray diffraction (XRD) of the composite nanostructures indicated that the as-prepared electrodes were well-aligned TiO₂ NTs with Zn_0.5Mg_0.5Fe₂O₄ nanoparticles. The enhanced absorption of the nanocomposite electrode under simulated sun light and visible light regions were observed. The photoelectrochemical performance of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ nanotube arrays showed excellent sensitive response to visible light. The surface-interface charge separation and transfer of photo-induced electrons and holes were also demonstrated by optical characterization. Meanwhile, the photoelectrochemical investigations clearly illustrated that the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ composite NTs had more effective photo-conversion capability than the unloaded TiO₂ NTs. The saturated photocurrent density of the doped electrode was about 6.5 fold and 4 fold as high as that of the TiO₂ NTs under the illumination of visible light and simulated sun light, respectively. In addition, the enhanced photoelectrocatalytic (PEC) ability of the as-prepared electrode was demonstrated in the degradation of toxic p-nitrophenol species. The much improved PEC activity can be attributed to both the visible-light photocatalytic activity of Zn_0.5Mg_0.5Fe₂O₄ and the heterostructure between Zn_0.5Mg_0.5Fe₂O₄ and TiO₂.

---

Introduction

Semiconductor-based photocatalysts have received considerable attention due to their potential in environmental pollution remediation and renewable energy applications.^1,2 TiO₂ nanostructures have in particular attracted tremendous interest due to their favourable photocatalytic (PC) activity, stability, anti-photocorrosion and non-toxicity.^3–5 Among the various nanostructures, well aligned TiO₂ nanotube arrays (NTs) are especially interesting and intriguing because of their large surface area and short diffusion length perpendicular to the charge collecting substrate, leading to lower recombination loss. As such, TiO₂ NTs have been developed for a range of applications including photocatalysis,^6–8 water splitting for hydrogen production,⁹ solar energy conversion,^10,11 lithium ion batteries¹² and gas sensors.¹³

However, TiO₂ is a semiconductor with wide band gap (3.0 and 3.18 eV for the rutile and anatase, respectively) only functioning under UV light irradiation (just about 4.5% of solar energy). Therefore, development of visible light driven photocatalysts based on TiO₂ NTs becomes a key challenge but holds a key to widespread applications of this type of photocatalysts. Various strategies have been used to extend the absorption band of TiO₂ to the visible light region such as metal ion doping,^14–17 nonmetal ion doping^18–20 and narrow-band-gap semiconductor composites.^21–25 The fact that the photon-generated electrons from the conduction band of the narrow band gap semiconductor could be injected into the conduction band of TiO₂ has been demonstrated, which not only can expand the spectral range of light absorption, but may also promote photon-excited electron–hole separation.^17,18

Recently, spinel ferrite (MeFe₂O₄, where Me usually represents a metal, such as Zn, Mg, Mn, Ni etc.) crystallites are considered good candidate for photocatalysis reaction in which Me²⁺ cations occupy tetrahedral sites.^26,27 Spinel ferrites offer the advantage of having a band gap capable of absorbing visible light, as well as the spinel crystal structure providing available extra catalytic sites by the virtue of crystal lattice.²⁸ A great deal of work has been reported about the combination of spinel ferrites with TiO₂ to enhance the PC activity. Especially, Zn_0.5Mg_0.5Fe₂O₄ with a spinel structure has been recently investigated due to its chemical stability and sensitivity to visible light.^29–31 Rekha Dom has synthesized Zn_0.5Mg_0.5Fe₂O₄ via microwave irradiation,²⁸ while S. Rahman has prepared ZnMg-ferrite nanoparticles using the co-precipitation method.³² Besides, both the conduction band and the valence band of Zn_0.5Mg_0.5Fe₂O₄ locate at higher wavelengths than those of TiO₂, resulting in an excellent separation of the electron–hole pairs.²⁷ Furthermore, the spinel ferrite properties of the dopants can enhance the separation efficiency of the photon-generated charges as well as their reactivity due to their internal electrostatic field.^27,33 Consequently, coupling TiO₂ NTs with Zn_0.5Mg_0.5Fe₂O₄ nanoparticles seems a promising approach to enhancing visible light absorption due to the special semiconductor structure and spatial crystal spinel structure of the Zn_0.5Mg_0.5Fe₂O₄ ferrites.

In this current work, we have synthesized visible-light-responsive Zn_0.5Mg_0.5Fe₂O₄ coupled TiO₂ NTs via an ultrasonically assisted electrodeposition strategy. The properties of the heterojunction between the Zn_0.5Mg_0.5Fe₂O₄ and TiO₂ NTs electrode and the photoelectrochemical properties of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode were determined and investigated using the photoelectrochemical techniques. In addition, the photoelectrocatalytic (PEC) activity of the electrodes was evaluated in detail by using the PEC degradation of toxic p-nitrophenol (PNP) species under simulated sun light and visible light irradiation conditions.

Experimental

Synthesis of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ composite electrode

Titanium foil (purity 99.6%, thickness 0.2 mm) was acquired from Beijing Academy of Steel Service, China. All the other chemicals were of analytical grade. The titanium foil was mechanically polished with sand papers of different grade (fineness) and then rinsed clean in an ultrasonic bath of deionized water for 10 min. The clean, polished titanium foil was subjected to chemically etching by immersing it in a mixture of HF and HNO₃ (HF/HNO₃/H₂O = 1 : 4 : 5 in volume) for 40 s. The titanium foil was rinsed again in deionized water to remove the mixed acid, followed by final cleaning by immersing it in an ethanol and deionized water solution in the ultrasonic bath for 15 min. The anodization was performed in a two-electrode cell with titanium foil as the working electrode and platinum foil as the counter electrode. The anodizing voltage was set at 20 V and anodized for 30 min in 0.2 wt% HF solution. After anodization, the sample was immediately rinsed with deionized water and dried at room temperature. The sample was then annealed at 773 K in air for 2 h with heating and cooling rates of 2 °C min⁻¹ to convert the amorphous phase to crystalline phase.

The deposition procedure was conducted in a three-electrode setup in an uniform cell, using the TiO₂ nanotube arrays electrode as the working electrode, a Pt foil as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode. The electrodeposition was carried out with a potential of −2.4 V for 20 min. Before electrodeposition, the sample was immersed in a mixture solution 50 mL containing 0.025 M (0.7425 g) Zn(NO₃)₂·6H₂O, 0.025 M (0.641 g) Mg(NO₃)₂·6H₂O and 0.1 M (4.04 g) Fe(NO₃)₃·9H₂O for 20 min in an ultrasonic bath at 40 kHz frequency, which drove the salt solution to diffuse through the nanochannels of TiO₂ and ensure that the inner walls of the pore channels were filled with the solution. Then the TiO₂ NTs electrode was diverted into a new medium that contained the supporting electrolyte (0.1 M Na₂SO₄, 80 mL). The deposition in this medium resulted in the deposition of Zn, Mg and Fe ions. After repeating the ultrasonication and deposition procedure several times, a desired amount of Zn, Mg and Fe deposition in the pores was achieved. Subsequently, the electrode was electrochemically oxidized in a 1 M KOH (70 mL) solution with a constant potential of 2.5 V for 5 min. Finally, the sample was rinsed with deionized water and dried in air. Then it was annealed by heating at 773 K for 2 h at muffle furnace, with a ramp of 2 °C min⁻¹.

Characterization

The morphology of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs was characterized using a field emission scanning electron microscope (SEM) (Quanta 450) with an accelerating voltage of 30.0 kV. The dopant concentration was characterized by energy-dispersive X-ray spectra (EDX) (Horiba 7593H), and EDX was performed to determine the elemental concentration distribution on the catalyst granules using Link Isis Series 300 software. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) images were taken on a FEI Tecnai G² F20 with an acceleration voltage of 200 and 300 kV. Chloroform solutions containing these nanocomposite samples were dropped on carbon-coated copper grids. The crystallinity of the sample was determined from X-ray diffraction patterns (XRD) using a diffractometer with Cu Kα radiation (Shimadzu Lab-X XRD-6000, source light at the wavelength (λ) of 0.15406 nm). The accelerating voltage and applied current were 30 kV and 30 mA, respectively. while the diffraction angles (2θ) was scanned from 20° to 60°. The absorption property was measured using UV-Vis diffuse reflectance spectra (DRS) (JASCO, UV-550) with a wavelength range of 200–700 nm. Photoluminescence (PL) spectra of the electrode surface was collected using a Hitachi F-4500 (excited at λ = 325 nm) fluorescence spectrophotometer in air at room temperature. X-ray photoelectron spectroscope (XPS) (PHI 5600 mode) was performed to examine the surface properties and composition of the sample. All the binding energies were calibrated by using the carbon (C 1s) 284.6 eV as a reference. The separation characteristics of photon-generated charge carriers was tested using a surface photovoltage (SPV) measurement system, which consisted of a monochromator (model Omni-λ 3005) and a lock-in amplifier (model SR830-DSP) with an optical chopper (model SR540) running at a frequency of 20 Hz.

PEC activity measurement

Photoelectrochemical measurement

Photocurrent density was measured using a CHI electrochemical analyzer (CHI 760C, Shanghai Chenhua, China) in a standard three-electrode configuration with Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs as the photoanode (an effective area of 5 cm²), a platinum foil as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode. A 0.01 M Na₂SO₄ (80 mL) solution was used as the electrolyte. The working electrode was irradiated with visible light (λ > 420 nm) through a UV-cutoff filter (Shanghai Seagull Colored Optical Glass Co. Ltd.) from a high-pressure xenon short arc lamp (a Phillips 500 W Xe lamp). The incident light intensity was measured with a radiometer (Photoelectric Instrument Factory Beijing Normal University, model FZ-A).

Photoelectrochemical experiment

The PEC reactions of PNP were carried out in a single photoelectrochemical compartment. The corresponding concentration of PNP species was obtained from diluting an analytical grade powder of PNP with deionized water. The as-prepared electrode serving as the photoanode and Pt foil as the cathode were placed in parallel in a cylindrical quartz reactor, and an SCE served as the reference electrode. All electrodes were connected to a CHI 760C EC analyzer. A bias potential applied on the photoanode was 0.6 V (vs. SCE) under visible light of 33.5 mW cm⁻² and simulated sun light of 36 mW cm⁻², respectively. All the experiments were performed with magnetic stirring, using a 0.01 M Na₂SO₄ solution as the electrolyte. The initial concentration of PNP in the aqueous solution (100 mL) for the degradation tests was 20 mg L⁻¹ under the simulated sun light and 5 mg L⁻¹ under the visible light, respectively. The concentration of PNP was determined using a UV 1100 spectrophotometer with the detection wavelength at 318 nm.

Results and discussion

Fig. 1a shows the top and cross-sectional (the inset) views of the TiO₂ NTs sample. Clearly, the self-organized TiO₂ nanotubes consist of vertically aligned compact nanotube arrays with a tube diameter of ∼90 nm, and length of ∼320 nm (the inset). Fig. 1b and c display the top and cross-sectional images of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs sample. The resulting highly ordered and vertically oriented nanotube arrays can be observed from Fig. 1b and c. It is obvious that the deposited nanoparticles not only entered the inner tubes but also surrounded the top of nanotube arrays without blocking the tube entrances. The nanotubes are still open at the top, indicating that the doping process of Zn_0.5Mg_0.5Fe₂O₄ did not destroy the structure of the nanotubes, which would induce the doped species in a much more uniform and compact manner within the inner-surface and outer-surface of the nanotube arrays. From the EDX spectrum of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs (Fig. 1d), it can be noticed that the as-prepared composite nanotube arrays contain Zn, Mg, Fe.

Fig. 1 Morphologies of TiO₂ nanotube arrays: (a) a typical top-view SEM image of TiO₂ NTs (inset is the cross-sectional image of TiO₂ NTs); (b) a top-view SEM image of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs; (c) a cross-sectional view of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs; (d) EDX of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs (inset table is the weight and atomic percentages of each element).

To further demonstrate the microstructure characteristics of Zn_0.5Mg_0.5Fe₂O₄ nanoparticles deposited on TiO₂ NTAs, the low (Fig. 2a–c) and high magnification (Fig. 2d) TEM images of Zn_0.5Mg_0.5Fe₂O₄/TiO₂ NTAs have been conducted. As exhibited by the low magnification TEM images in Fig. 2a and b, Zn_0.5Mg_0.5Fe₂O₄ nanoparticles are deposited on the tubes entrance while causing no damage of nanotubes. Seen from Fig. 2c, it can be found that Zn_0.5Mg_0.5Fe₂O₄ nanoparticles with a diameter of 20–30 nm are deposited into the tube. Fig. 2d depicts the HRTEM image of the selected area, where the lattice fringes of 0.253 nm and 0.351 nm correspond to the reflections from the (311) planes of spinel Zn_0.5Mg_0.5Fe₂O₄ (JCPDS no. 04-002-5442)³⁴ and the (101) plane of anatase TiO₂ (JCPDS no. 21-1272),⁶ respectively. The TEM testing results confirm that spinel ferrite Zn_0.5Mg_0.5Fe₂O₄ nanoparticles are successfully deposited to TiO₂ NTAs.

Fig. 2 TEM (a–c) and HRTEM (d) images of the Zn_0.5Mg_0.5Fe₂O₄/TiO₂ NTAs nanocomposite electrode.

In order to identify the crystal phases of TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs, XRD measurements was conducted under specified conditions. Fig. 3 shows the XRD patterns of TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs, respectively. The diffraction peaks at 35.2°, 38.4°, 40.3° and 53.1° correspond to (100), (002), (101) and (102) crystal faces of the Ti sheet (JCPDS no. 07-6265). Obviously, TiO₂ NTs annealed at 773 K possesses the crystal structure of anatase (2θ = 25.3°, 48.1°. JCPDS no. 07-6175). After the deposition of Zn_0.5Mg_0.5Fe₂O₄, four additional diffraction peaks are identified at 2θ = 30.1°, 35.4°, 43.1° and 57°, which are in accordance with the (220), (331), (400) and (511) planes of spinel Zn_0.5Mg_0.5Fe₂O₄ (JCPDS no. 04-001-9289) respectively.³⁵ The sharp peak near 35.3° in Fig. 3b is attributed to the overlap and accumulation of 35.2° from anatase and 35.4° from spinel ferrite. It should be noted that isolated MgO, Fe₂O₃ and ZnO phases are not found in the obtained Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs sample. The XRD results indicate that the composite nanostructures grown on electrodes prepared by the ultrasonically assisted electrodeposition approach are composed of the desired spinel Zn_0.5Mg_0.5Fe₂O₄ and anatase TiO₂ crystalline structures.

Fig. 3 XRD patterns of Ti sheet (a), TiO₂ nanotube arrays (b), and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs (c), (ZMF in (c) represents Zn_0.5Mg_0.5Fe₂O₄).

Fig. 4a shows the UV-Vis DRS of the TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs. It can be seen that, in the visible light region, the absorption intensity of the sample Zn_0.5Mg_0.5Fe₂O₄@TiO₂ is dramatically increased, which clearly reveals that Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs are more sensitive to the visible light than that of the non-doped NTs, therefore, it follows that the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode possesses a much improved photoelectrochemical capability under visible light irradiation. At the same time, an obviously red shift of the absorption threshold toward the visible light region has been successfully observed for the Zn_0.5Mg_0.5Fe₂O₄-doped sample. Two characteristic absorption peaks of TiO₂ at about 410 nm and 540 nm corresponded to the absorption of the trapped hole and the trapped electron, respectively, which result from the sub-bandgap of surface defects on TiO₂ nanotubes. According to literatures,^36,37 the hole traps or deep electron-acceptor type defects showed a peak absorption at around λ = 410 nm. The electron traps exhibited another peak absorption around λ = 540 nm, which can be attributed to the oxygen vacancies related sub-bandgap states of the TiO₂ NTs. Fig. 4b shows the (αhν)^1/2 versus hν curve of TiO₂ and Zn_0.5Mg_0.5Fe₂O₄ on TiO₂ NTs coming from the equation (αhν)^1/2 = k(hν − E_g), where α, h, ν and E_g represent the adsorption coefficient, Planck constant, light frequency and band energy, respectively.^38,39 For TiO₂ NTs, the absorption onset is at approximately 387 nm, which corresponds well to the band gap 3.18 eV of anatase TiO₂. The absorption of TiO₂ NTs in the visible range could be assigned to the sub-band gap states of the TiO₂ NTs.⁴⁰ After the deposition of Zn_0.5Mg_0.5Fe₂O₄ nanoparticles, the absorption threshold shifts to 570 nm. The band gap is estimated to be 2.0 eV.

Fig. 4 (a) UV-Vis DRS of TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs; (b) Tauc's plots to determine the band gaps for TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs.

Consequently, the DRS spectra suggest that the Zn_0.5Mg_0.5Fe₂O₄-doped nanotube arrays might exhibit a higher visible light response than the non-doped NTs. arrays might exhibit a higher visible light response than the non-doped NTs.

In order to reveal the impact of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs photo-induced charge separation, migration and combinations,⁴¹ PL spectra of the samples are characterized. Fig. 5 presents the TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄ @TiO₂ NTs PL spectra. The excitation wavelength of the samples tested was 325 nm. TiO₂ NTs and doped TiO₂ NTs in the vicinity of 400 nm exhibit a strong emission peak.

Fig. 5 PL spectra of TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs.

Two weak emission peaks also exist around 454 nm and 529 nm, respectively. The emission peak around 400 nm is attributed to the band edge emission of TiO₂ NTs (intrinsic luminescence), corresponding to electron transitions between the emission band.⁴² The two weak peeks around 454 nm and 529 nm are ascribed to TiO₂ NTs with edges of free excitons and bound excitons. These PL signals are mainly due to oxygen vacancies and defects generated on the nanoparticle surface.^43,44 Compared with the PL spectra of TiO₂ NTs and doped TiO₂ NTs, the PL intensity of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs is found to be significantly reduced. Since the PL spectrum results from the recombination of electrons and holes, a lower luminous intensity means less recombination of electrons and holes.

The experimental results clearly indicate that the recombination ratio of photo-generated electron–hole pairs is much lowered due to the loading of Zn_0.5Mg_0.5Fe₂O₄.⁴⁵

To confirm the states of elements and composition information in the nanotubular samples, XPS was employed to reveal the valence states and surface chemical compositions of the composite nanotube arrays. Elements of Ti, Mg, Zn, Fe, O and adventitious carbon are totally proved to be presented in the composite (Fig. 6a). Carbon is ascribed to the adventitious hydrocarbon from the XPS instrument itself. In Fig. 6b, two peaks for the Ti 2p are observed at binding energy 464.8 eV and 459.2 eV, which are assigned to Ti 2p_1/2 and Ti 2p_3/2 respectively. These values are in good agreement with the XPS data known for Ti⁴⁺ in the pure anatase titania form⁴⁶ in the corresponding literature. It seems that loading Zn_0.5Mg_0.5Fe₂O₄ species did not have any effect on the position of Ti 2p peak. As shown in Fig. 6c, XPS peaks of Zn 2p located at 1044.4 eV and 1022.8 eV can be assigned to the spectra of Zn 2p_1/2 and Zn 2p_3/2, suggesting that the composite electrode contains Zn²⁺.⁴⁷ In Fig. 6d, the peaks located at 710.6 eV and 724.4 eV can be attributed to Fe 2p_3/2 and Fe 2p_1/2 for Fe³⁺, respectively, which reveal the oxidation state of Fe³⁺ in the nanocomposites.⁴⁸ In addition, in Fig. 6e, it can be found that a sharp peak around 1303.8 eV, which matches well with Mg 1s, indicating that Mg exists in Mg²⁺ oxidation state in the composite electrode.³³ As shown in Fig. 6f, O 1s peak is asymmetric suggesting at least two kinds of oxygen species are present in and near the surface region. Meanwhile, the XPS spectra of O 1s of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs are fitted using Gaussian–Lorentzian peak shapes, which can be devolved into three small peaks. The peak located at 532.4 eV is assigned to the adsorbed oxygen. The other two peaks located at 529.4 eV and 530.1 eV are attributed to the O element in TiO₂ and Zn_0.5Mg_0.5Fe₂O₄, respectively.⁴⁹

Fig. 6 The whole XPS (a), Ti 2p (b), Fe 2p (c), Zn 2p (d), Mg 1s (e) and O 1s (f) XPS of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ composite NTs.

To further demonstrate the effect of Zn_0.5Mg_0.5Fe₂O₄ on expanding the active spectral range of doped TiO₂ NTs, SPV characterization of the as-prepared electrodes was conducted (Fig. 7). TiO₂ NTs electrode presents a response in the UV range from 300 to 380 nm, which is ascribed to the electron transition from valence band to conduction band of TiO₂ (O_2p → Ti_3d).⁵⁰ However, the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode displays a stronger SPV signal than that of the TiO₂ NTs electrode. The modified electrode exhibits a red shift of the spectral band towards visible region, which is also confirmed by the UV-Vis DRS characterization. In view of that the strong SPV signal has direct correlation with effective separation of photo-generated charge carriers,⁵¹ the presence of Zn_0.5Mg_0.5Fe₂O₄ would favor the photo-induced charge carrier separation and surface-interface transfer, greatly enhancing the PC and PEC activity of TiO₂ NTs electrode.

Fig. 7 SPV spectra of the as-prepared TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs samples.

Photoelectrochemical performance

The photoelectrochemical performance of TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs was investigated under visible light and simulated sun light irradiation. The photocurrent densities as a function of the applied potential of the electrodes in a 0.01 M Na₂SO₄ solution are presented in Fig. 8a and c. The dark current densities are low and negligible under visible light and simulated sun light irradiation. However, TiO₂ still displays a noticeable response while the photocurrent density of doped TiO₂ NTs dramatically rises under the visible light. The increase and subsequent decrease of photocurrent density is believed to be due to the fact that the space charge region of the heterojunction promotes the separation of photo-induced carriers by the internal electrostatic field in the heterojunction region. As a consequence, more charge carriers would participate in reactions instead of recombination.⁵² However, this internal electrostatic field of the junction region would be weakened with increasing the applied bias potential because the thickness of the space charge layer decreased with the bias voltage. As a result, less charge carriers could be separated, and the photocurrent density decreases accordingly.^53,54 The saturated photocurrent density of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode, at 0.13 mA cm⁻² is about 6.5 folds as high as that of TiO₂ NTs at 0.02 mA cm⁻² under visible light irradiation. In the meantime, the photocurrent density of the doped TiO₂ NTs electrode is more than 4 folds as high as that of the TiO₂ NTs (1.75 versus 0.4 mA cm⁻²) under simulated sun light irradiation. It may therefore be concluded that Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs has a much higher conversion capability of visible light than TiO₂ NTs, at a 10 folds increase, and the photo-generated carrier-separation efficiency has been improved significantly after being modified with Zn_0.5Mg_0.5Fe₂O₄ nanoparticles. As shown in Fig. 8b and d, after loading Zn_0.5Mg_0.5Fe₂O₄ nanoparticles, the photo-conversion efficiency has been improved under simulated sun light and visible light irradiation. The photo-conversion efficiency (η) of light energy to chemical energy in the presence of a bias potential is calculated as follows.⁶
where, j_p means the photocurrent density (mA cm⁻²), j_pE⁰_rev is the total power output; j_p|E_app| represents the electrical power input; and I₀ is the power density of the incident light (mA cm⁻²). E⁰_rev is the standard reversible potential (which is 1.23 V for the water-splitting reaction when pH value reaches 0), and |E_app| is the absolute value of the applied potential E_app, which can be calculated as E_app = E_meas − E_aoc, where E_meas is the electrode potential (vs. SCE) of the working electrode at which the photocurrent is measured under illumination and E_aoc is the electrode potential (vs. SCE) of the same working electrode under open-circuit conditions, under the same condition.

Fig. 8 I–V characteristics of TiO₂ and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs under visible light (I₀ = 33.5 mW cm⁻², λ > 420 nm, 0.6 V vs. SCE in 0.01 M Na₂SO₄) and simulated sun light irradiation (I₀ = 36 mW cm⁻², 0.6 V vs. SCE in 0.01 M Na₂SO₄) (a) and (c); and photoconversion efficiency as a function of applied bias potential under visible light and simulated sun light irradiation (b) and (d).

In order to study the properties of the electrodes and their interactions with the solution, the EIS measurements of the TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs are carried out, which cover the frequencies of 10⁵ to 10⁻² Hz intervals by an amplitude of 10 mV at an open circuit potential under dark and visible light irradiation.⁴⁹ By controlling the electrode current or potential, the EIS curves of the as-prepared electrode vary by slight amplitude sinusoidal curve with time, while measuring the potential or current variation with time of the electrode.⁵⁵ Fig. 9 exhibits the EIS results of the TiO₂ NTs and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrodes under dark and visible illumination in the 0.01 M Na₂SO₄ electrolyte, with the bias voltage set to 0.3 V. The results in Fig. 9 suggest that without any light irradiation, the Nernst curve of TiO₂ NTs becomes close to a straight line, indicating that the TiO₂ NTs electrode response is poor without any light irradiation. The impedance of the radius of the TiO₂ NTs electrode decreases sharply after Zn_0.5Mg_0.5Fe₂O₄ loading or with visible light irradiation, which means that the composite material has a good response to visible light region. The EIS measurement results illustrate that Zn_0.5Mg_0.5Fe₂O₄ loading effectively improves the separation of photon-generated electron–hole pairs and accelerates the photon-generated charge transfer between Zn_0.5Mg_0.5Fe₂O₄ and TiO₂ NTs interface. It is concluded that the doping Zn_0.5Mg_0.5Fe₂O₄ nanoparticles effectively accelerate the transfer of photon-generated charge carriers.

Fig. 9 EIS Nyquist plots of TiO₂ NTs electrode and Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode under dark and visible light irradiation (λ > 420 nm, I₀ = 33.5 mW cm⁻²) (the bias potential is 0.3 V (vs. SCE) in a 0.01 M Na₂SO₄ aqueous solution).

Degradation of PNP

To evaluate the PEC activity of the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode, a series of experiments for PNP degradation were carried out. After irradiation for 120 min under simulated sun light, the concentration of PNP was almost unchanged in the direct photocatalytic (DP) degradation without any photocatalyst, while 10.5% of PNP was degraded in the electrochemical process (EC) (Fig. 10a). The degradation efficiency of PNP by PC degradation was 36% with the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ composite nanotube array electrode. When a bias potential of 0.6 V (vs. SCE) was applied, the degradation efficiency increased to about 69% for the Zn_0.5Mg_0.5Fe₂O₄ loaded electrode, an enhanced of PNP than heterojunction network catalyst of Cu₂O/TiO₂.⁵⁶ Compared with the non-doped TiO₂ NTs, the composite electrodes afford a 2 folds increase in the efficiency in PNP degradation. In the meantime, trends can be observed in Fig. 10c, where the EC, DP, PC and PEC degradation efficiency of PNP by Zn_0.5Mg_0.5Fe₂O₄ are 9.3%, 10.3%, 14.8% and 59%, respectively, when illuminated by visible light while the PEC degradation of PNP is just 21% by TiO₂ NTs alone. Judging from the increased degradation efficiency, there exists an obvious synergistic effect between the EC process and the PC process (under both simulated sun light irradiation and visible light irradiation, respectively). The enhanced PEC activity of the composite electrode could be attributed to the comprehensive effect between the lowered electron–hole recombination rate by the applied bias potential and the wider spectral response promoted by the Zn_0.5Mg_0.5Fe₂O₄ nanoparticles.^6,56 Thus the applied bias potential effectively inhibits the recombination of photo-generated electron–hole pairs and prolongs the life of the photo-generated charge carriers, hence yielding a better photoelectrochemical performance. As shown in Fig. 10b and d, the kinetics of PNP degradation processes in EP, DP, PC, and PEC by Zn_0.5Mg_0.5Fe₂O₄ doped electrode and PEC by TiO₂ NTs follows a pseudo-first-order kinetics, as described in Table 1.

Fig. 10 The PNP concentration vs. time plots in the PNP degradation tests with nanocomposite electrodes in different processes under simulated sun light irradiation: (a) (I₀ = 36 mW cm⁻², 0.6 V vs. SCE, C₀ = 20 mg L⁻¹) and visible light irradiation (c) (I₀ = 33.5 mW cm⁻², λ > 420 nm, 0.6 V vs. SCE, C₀ = 5 mg L⁻¹); and kinetic constants and regression coefficients of degradation for PNP obtained under simulated sun light irradiation (b) and visible light irradiation (d).

Table 1 Kinetic constants and regression coefficients for PNP degradation using different processes under simulated sun light (a) and visible light irradiation (b)

Processes	Kinetic constants (min^−1)	R^2
(a)
EC	3 × 10^−4	0.9441
DP	9 × 10^−4	0.9794
PC	3.7 × 10^−3	0.9912
PEC	8 × 10^−3	0.9906
TiO2	3.8 × 10^−3	0.9872
(b)
EC	4 × 10^−5	0.9873
DP	5 × 10^−5	0.9882
PC	8 × 10^−4	0.9933
PEC	2.8 × 10^−3	0.9941
TiO2	8 × 10^−4	0.9713

---

To further investigate the active species involved in PEC degradation, we have conducted the capture of the active species in PNP degradation to determine which species play a major role in PNP degradation. As exhibited in Fig. 11, the PEC degradation of PNP was inhibited to some extent by the addition of t-BuOH (hydroxyl radicals scavenger) under simulated sun light irradiation, indicating that the hydroxyl radicals were confirmed as active oxidative species in the PEC process. However, it was intensively suppressed when EDTA (holes scavenger) was introduced. This result indicates that photogenerated holes were the main active oxidizing species involved in the PEC process compared with other species. The direct holes transfer mainly governed the PC process. Meanwhile, ˙O₂⁻ might also serve as active species in this reaction. N₂ was used to confirm the effect of O₂ and the results are shown in Fig. 11. Under the anoxic condition, the degradation of PNP was not completely inhibited, indicating that ˙O₂⁻ was active species but not the main one. In summary, the photogenerated holes account for the photo activity most compared with ˙OH and ˙O₂⁻ in PEC degradation of PNP.

Fig. 11 Plots of photogenerated active species trapped in the system of photodegradation of PNP by Zn_0.5Mg_0.5Fe₂O₄/TiO₂ NTAs under simulated light irradiation (I₀ = 36 mW cm⁻², 0.6 V vs. SCE and C₀ = 20 mg L⁻¹ in 0.01 M Na₂SO₄).

The stability and recyclability of the as-prepared electrode was evaluated by repeating the PNP degradation under simulated sun light irradiation several times. Before each repeating experiment, the composite electrode was ultrasonically cleaned with deionized water and each new PEC degradation cycle was carried out by introducing a fresh PNP solution and Na₂SO₄ electrolyte under the same conditions. As shown in Fig. 12, the efficiency of PNP degradation shows no obvious decline in the PEC activity of Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs after five recycles. Therefore, it is reasonably concluded that the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode exhibits fair stability and can be optimized as a suitable candidate for PEC purification of PNP polluted water.

Fig. 12 PEC degradation of PNP by the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs electrode over five consecutive cycles under simulated sun light illumination.

Mechanism of enhanced PEC activity

To gain a further insight into the enhanced PEC activity, the charge transfer behavior during PNP degradation is illustrated in Fig. 13. The involved photoelectrocatalytic reactions of PNP degradation over Zn_0.5Mg_0.5Fe₂O₄@TiO₂ nanotube arrays have been described as follows (herein, ZMF represents Zn_0.5Mg_0.5Fe₂O₄). To calculate the band structures of Zn_0.5Mg_0.5Fe₂O₄, periodic density functional calculations (DFT) computations were performed using a plane-wave method implemented in the Cambridge Sequential Total Energy Package (CASTEP) code.^57,58 It can be obtained that the band gap of Zn_0.5Mg_0.5Fe₂O₄ is approximately 1.9 eV, in good agreement with DRS experimental values 2.0 eV (the conduction band and valence band structure of Zn_0.5Mg_0.5Fe₂O₄ can be seen in ESI†). As shown in Fig. 13, under visible light irradiation, Zn_0.5Mg_0.5Fe₂O₄ with a narrow band gap can be easily excited to generate electron–hole pairs while TiO₂ with a band gap of 3.2 eV can not be excited. The photo-generated electrons can be forced to transfer from the conduction band of Zn_0.5Mg_0.5Fe₂O₄ to TiO₂ under an applied bias potential of 0.6 V, leaving more holes in the valence band of Zn_0.5Mg_0.5Fe₂O₄, significantly reducing the charge recombination. However, the holes spontaneously transfer from TiO₂ NTs to the valence band of Zn_0.5Mg_0.5Fe₂O₄ nanoparticle. This separation effectively prevent the Zn_0.5Mg_0.5Fe₂O₄ react with H₂O or hydroxyl ions (OH⁻) to form hydroxyl radicals (˙OH). The accumulated electrons in the conduction band of TiO₂ are transferred to the adsorbed O₂ on the TiO₂ surface to form superoxide radical anions (˙O₂⁻), which are then reduced to ˙OH. Eventually, the photo-generated active oxidizing species such as ˙OH, ˙O₂⁻ and h⁺ are involved in the self-sensitized photoelectrocatalysis degradation of PNP.

ZMF + visible light → ZMF(h⁺) + ZMF(e⁻)

ZMF(e⁻) + TiO₂ → ZMF + TiO₂(e⁻)

TiO₂(e⁻) + O₂ → TiO₂ + ˙O₂⁻

˙O₂⁻ + H⁺ → ˙OOH → H₂O₂

H₂O₂ + ˙O₂⁻ → ˙OH + OH⁻

TiO₂(e⁻) + external electrostatic field → external circuit → counter electrode

ZMF(h⁺) + OH⁻ → ZMF + ˙OH

PNP + ˙OH, ˙O₂⁻, h⁺ → intermediate + H₂O + CO₂

Fig. 13 Schematic illustration of the mechanism of photocatalytic degradation of PNP using Zn_0.5Mg_0.5Fe₂O₄@TiO₂ NTs as the photocatalyst under visible light irradiation (λ > 420 nm, I₀ = 33.5 mW cm⁻²).

Conclusions

A novel Zn_0.5Mg_0.5Fe₂O₄@TiO₂ composite-nanotube-array photo-electrode was successfully synthesized using an ultrasonically assisted electrodeposition technique. The electrodeposition strategy effectively promoted the deposition of the Zn_0.5Mg_0.5Fe₂O₄ nanoparticles within self-organized and vertically oriented TiO₂ nanotube arrays while minimizing the clogging of the nanotube entrances. The PEC properties demonstrated that the composite nanotube arrays showed typical heterojunction characteristics. The loading of Zn_0.5Mg_0.5Fe₂O₄ nanoparticles significantly extended the response of the TiO₂ nanotube arrays to the visible light region. The composite nanotube arrays showed a significantly higher photocurrent density and improved photo-conversion efficiency due to better charge separation and collection efficiency under simulated sun light and visible light illumination. In addition, the composite electrode was found to possess excellent PEC activity for the degradation of PNP as compared with the non-doped TiO₂ electrode under both simulated sun light and visible light illumination. Consequently, the Zn_0.5Mg_0.5Fe₂O₄@TiO₂ heterostructure photoelectrode described herein may be used in a wide range of applications including photocatalysis.

Acknowledgements

This work received partial financial support from the National Nature Science Foundation of China (no. NSFC21377015, 21207015), the Major State Basic Research Development Program of China (973 Program) (no. 2011CB936002) and the Key Laboratory of Industrial Ecology and Environmental Engineering, China Ministry of Education.

Notes and references

1. X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503–6570.
2. H. Kisch, Angew. Chem., Int. Ed., 2013, 52, 812–847.
3. A. L. Linsebigler, G. Lu and J. T. Yates Jr, Chem. Rev., 1995, 95, 735–758.
4. S. G. Kumar and L. G. Devi, J. Phys. Chem. A, 2011, 115, 13211–13241.
5. K. Lee, D. Kim, P. Roy, I. Paramasivam, B. I. Birajdar, E. Spiecker and P. Schmuki, J. Am. Chem. Soc., 2010, 132, 1478–1479.
6. A. Zhu, Q. Zhao, X. Li and Y. Shi, ACS Appl. Mater. Interfaces, 2014, 6, 671–679.
7. T. Li, X. Li, Q. Zhao, Y. Shi and W. Teng, Appl. Catal., B, 2014, 156–157, 362–370.
8. A. E. R. Mohamed and S. Rohani, Energy Environ. Sci., 2011, 4, 1065–1086.
9. R. M. Navarro, M. C. Alvarez-Galván, J. A. V. de la Mano, S. M. Al-Zahrani and J. L. G. Fierro, Energy Environ. Sci., 2010, 3, 1865–1882.
10. C. X. Kronawitter, L. Vayssieres, S. Shen, L. Guo, D. A. Wheeler, J. Z. Zhang, B. R. Antoun and S. S. Mao, Energy Environ. Sci., 2011, 4, 3889–3899.
11. A. Lamberti, A. Sacco, S. Bianco, D. Manfredi, F. Cappelluti, S. Hernandez, M. Quaglio and C. F. Pirri, Phys. Chem. Chem. Phys., 2013, 15, 2596–2602.
12. L. Xue, Z. Wei, R. Li, J. Liu, T. Huang and A. Yu, J. Mater. Chem., 2011, 21, 3216–3220.
13. L. Sun, S. Zhang, X. W. Sun, X. Wang and Y. Cai, Langmuir, 2010, 26, 18424–18429.
14. W. Teng, X. Y. Li, Q. D. Zhao and G. H. Chen, J. Mater. Chem. A, 2013, 1, 9060–9068.
15. W. Wang, M. Tian, A. Abdulagatov, S. M. George, Y. C. Lee and R. Yang, Nano Lett., 2012, 12, 655–660.
16. J. Choi, H. Park and M. R. Hoffmann, J. Phys. Chem. C, 2009, 114, 783–792.
17. M. A. Hashemian, E. Palacios, I. I. Nedrygailov, D. Diesing and E. Karpov, ACS Appl. Mater. Interfaces, 2013, 5, 12375–12379.
18. W. Tu, Y. Zhou, Q. Liu, S. Yan, S. Bao, X. Wang, M. Xiao and Z. Zou, Adv. Funct. Mater., 2013, 23, 1743–1749.
19. H. U. Lee, S. C. Lee, S. H. Choi, B. Son, S. J. Lee, H. J. Kim and J. Lee, Appl. Catal., B, 2013, 129, 106–113.
20. A. Ghicov, J. M. Macak, H. Tsuchiya, J. Kunze, V. Haeublein, L. Frey and P. Schmuki, Nano Lett., 2006, 6, 1080–1082.
21. M. Wang, L. Sun, Z. Lin, J. Cai, K. Xie and C. Lin, Energy Environ. Sci., 2013, 6, 1211.
22. G. Li, L. Wu, F. Li, P. Xu, D. Zhang and H. Li, Nanoscale, 2013, 5, 2118–2125.
23. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2008, 130, 4007–4015.
24. Q. Kang, S. Liu, L. Yang, Q. Cai and C. A. Grimes, ACS Appl. Mater. Interfaces, 2011, 3, 746–749.
25. W.-T. Sun, Y. Yu, H.-Y. Pan, X.-F. Gao, Q. Chen and L.-M. Peng, J. Am. Chem. Soc., 2008, 130, 1124–1125.
26. K. Thummer, M. Chhantbar, K. Modi, G. Baldha and H. Joshi, J. Magn. Magn. Mater., 2004, 280, 23–30.
27. E. Casbeer, V. K. Sharma and X.-Z. Li, Sep. Purif. Technol., 2012, 87, 1–14.
28. R. Dom, R. Subasri, K. Radha and P. H. Borse, Solid State Commun., 2011, 151, 470–473.
29. M. Gaudon, N. Pailhé, A. Wattiaux and A. Demourgues, Mater. Res. Bull., 2009, 44, 479–484.
30. K. A. Mohammed, A. D. Al-Rawas, A. M. Gismelseed, A. Sellai, H. M. Widatallah, A. Yousif, M. E. Elzain and M. Shongwe, Phys. B, 2012, 407, 795–804.
31. A. Xia, S. Liu, C. Jin, L. Chen and Y. Lv, Mater. Lett., 2013, 105, 199–201.
32. S. Rahman, K. Nadeem, M. Anis-ur-Rehman, M. Mumtaz, S. Naeem and I. Letofsky-Papst, Ceram. Int., 2013, 39, 5235–5239.
33. Y. Hou, F. Zuo, A. Dagg and P. Feng, Angew. Chem., 2013, 52, 1248–1252.
34. K. Mukherjee and S. Majumder, Nanotechnology, 2010, 21, 255504.
35. K. Mukherjee and S. Majumder, Sens. Actuators, B, 2013, 177, 55–63.
36. Z. Wu, Y. Wang, L. Sun, Y. Mao, M. Wang and C. Lin, J. Mater. Chem. A, 2014, 2, 8223–8229.
37. S. Kurian, H. Seo and H. Jeon, J. Phys. Chem. C, 2013, 117, 16811–16819.
38. M. Paulose, K. Shankar, S. Yoriya, H. E. Prakasam, O. K. Varghese, G. K. Mor, T. A. Latempa, A. Fitzgerald and C. A. Grimes, J. Phys. Chem. B, 2006, 110, 16179–16184.
39. K. Shankar, G. K. Mor, A. Fitzgerald and C. A. Grimes, J. Phys. Chem. C, 2007, 111, 21–26.
40. H. E. Prakasam, K. Shankar, M. Paulose, O. K. Varghese and C. A. Grimes, J. Phys. Chem. C, 2007, 111, 7235–7241.
41. N. Serpone, D. Lawless and R. Khairutdinov, J. Phys. Chem., 1995, 99, 16646–16654.
42. V. Salas, E. Olias, A. Barrado and A. Lazaro, Sol. Energy Mater. Sol. Cells, 2006, 90, 1555–1578.
43. Y. Lei, L.-D. Zhang, G.-W. Meng, G.-H. Li, X. Zhang, C. Liang, W. Chen and S. Wang, Appl. Phys. Lett., 2001, 78, 1125–1127.
44. T. Tachikawa and T. Majima, Langmuir, 2009, 25, 7791–7802.
45. T. Tachikawa, N. Wang, S. Yamashita, S. C. Cui and T. Majima, Angew. Chem., Int. Ed., 2010, 49, 8593–8597.
46. Y. Hou, X.-Y. Li, Q.-D. Zhao, X. Quan and G.-H. Chen, Adv. Funct. Mater., 2010, 20, 2165–2174.
47. Y.-Q. Chu, Z.-W. Fu and Q.-Z. Qin, Electrochim. Acta, 2004, 49, 4915–4921.
48. Y. Sharma, N. Sharma, G. Rao and B. Chowdari, Electrochim. Acta, 2008, 53, 2380–2385.
49. M. Wang, L. Sun, J. Cai, P. Huang, Y. Su and C. Lin, J. Mater. Chem. A, 2013, 1, 12082.
50. Y. Mizukoshi, N. Ohtsu, S. Semboshi and N. Masahashi, Appl. Catal., B, 2009, 91, 152–156.
51. Q. Li and J. K. Shang, Environ. Sci. Technol., 2009, 43, 8923–8929.
52. Y. Chen, J. C. Crittenden, S. Hackney, L. Sutter and D. W. Hand, Environ. Sci. Technol., 2005, 39, 1201–1208.
53. Y. Hou, F. Zuo, A. Dagg and P. Feng, Nano Lett., 2012, 12, 6464–6473.
54. H. Yu, X. Quan, S. Chen and H. Zhao, J. Phys. Chem. C, 2007, 111, 12987–12991.
55. X. Zhao, T. Xu, W. Yao, C. Zhang and Y. Zhu, Appl. Catal., B, 2007, 72, 92–97.
56. L. Yang, S. Luo, Y. Li, Y. Xiao, Q. Kang and Q. Cai, Environ. Sci. Technol., 2010, 44, 7641–7646.
57. L. Sun, L. Xiang, X. Zhao, C.-J. Jia, J. Yang, Z. Jin, X. Cheng and W. Fan, ACS Catal., 2015, 5, 3540–3551.
58. M. Nolan, ACS Appl. Mater. Interfaces, 2012, 4, 5863–5871.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra06443k

---