Abstract

A novel method for decoration of anodic TiO2 nanotube array films (NAFs) with Cu2O nanoparticles has been reported. The method is based on the reduction of Cu(II) in a mixture of ethylene glycol and N,N-dimethylformamide at 120°C for 16 h, where the resulting Cu2O can heterogeneously nucleate and grow on TiO2 NAFs. The nanosized Cu2O is found to be well dispersed on the wall of TiO2 nanotubes without blocking the nanotube, a commonly observed phenomenon in the case of deposition of Cu2O via electrochemical method. The amount of Cu2O deposited on the TiO2 NAFs can be varied by adjusting the concentration of Cu(II) in the organic solution. UV-vis spectra measurement indicates that the decoration of TiO2 NAFs with Cu2O nanoparticles greatly improves their ability to respond to visible light. By examining the photocurrent and photodegradation of methyl orange under simulated sunlight, it is found that these Cu2O-decorated TiO2 NAFs show much more photoactive in comparison with the as-prepared TiO2 NAFs.

1. Introduction

Because of their large aspect ratio and high specific surface area, materials with one-dimensional (1D) structures, for example, nanotubes and nanowires, often exhibit different performances compared to the bulk counterparts. The unique properties of these 1D nanostructures have shown potential applications in many fields, such as electronics, catalysis, data storage, optics, and sensors [13]. 1D TiO2 nanostructures are of great scientific and technical interest because they exhibit excellent photocatalytic activities [47]. Over the past decade, great attention has been paid into the synthesis and application of TiO2 nanotube arrays prepared by anodic oxidation of Ti metal in F-containing solutions [711]. The highly ordered nanotube arrays not only possess high surface area, but also provide an efficient transport channel for photogenerated electrons [12]. Furthermore, unlike the powder-typed TiO2 photocatalysts which often need be immobilized onto solid substrates for practical application [13, 14], the TiO2 nanotube arrays are grown on Ti substrates and thus the formed TiO2 nanotube array films (NAFs) can be directly used as photoanodes for photoinduced redox reactions such as water splitting [15] and decomposition of harmful compounds [7].

However, TiO2 nanotube arrays possess a wide band gap (~3.2 eV) and thus only respond well to ultraviolet light, which is a great hindrance to their use under sunlight. To extend their light-response scope from ultraviolet to visible light region, a common approach is postdecoration of TiO2 NAFs with narrow band gap semiconductors, for example, Cu2O [1620], Fe2O3 [21] and CdS [22]. When TiO2 NAFs are decorated with Cu2O, a p-type semiconductor with a direct band gap of ~2.2 eV, electrons excited under visible light may transfer from the conduction band of Cu2O to that of TiO2 since the conduction band (CB) edge for Cu2O is much higher than that of TiO2 [23]. As a result, the recombination probability of the photoexcited electrons and holes will be reduced, leading to a great improvement in photocatalytic activity. So far, the methods for the decoration of TiO2 nanotube arrays with Cu2O mainly include electrodeposition [1618], sonoelectrochemical deposition [19], and photocatalytic reduction [20]. In the present work, we report a new method for loading of Cu2O nanoparticles onto TiO2 NAFs. The method is based on the reduction of Cu (II) in a mixture of ethylene glycol (EG) and N,N-dimethylformamide (DMF), where the resulting Cu2O can heterogeneously nucleate and grow on TiO2 NAFs. In comparison with commonly used electrodeposition (including sonoelectrochemical deposition), the size of Cu2O is small, and the nanosized Cu2O is well dispersed on the wall of TiO2 nanotubes without blocking the nanotube.

2. Experimental Details

2.1. Synthesis of the Films

Ti foils were cut into pieces (7.2 cm × 1.7 cm × 0.4 mm), polished with abrasive paper, and then washed with deionized water. The polished Ti pieces were degreased in a mixed solution of NaOH and Na2CO3 (the ratio of NaOH : Na2CO3 : H2O by weight is 5 : 2 : 100, resp.) at 85°C for 1.5 h, and then washed with deionized water. Before anodic oxidization, one side of the pretreated Ti piece was sealed with epoxy resin, and then etched in a 10 wt% HF aqueous solution at room temperature for about 20 s, followed by washing with deionized water. The anodic oxidization of the Ti piece was conducted in an EG solution containing KF (0.7 wt%) and H2O (1.8 wt%) at ~25°C, where a Cu plate was used as cathode and a constant voltage of 50 V was applied between two electrodes. The anodic oxide layer was formed by a three-step method. Firstly, the Ti piece was anodized for 2 h, and then the grown oxide layer was removed by an adhesive tape. Secondly, the above procedure was repeated. Finally, the Ti piece was reanodized under the same conditions for 1 h. After anodization, the sample was washed thoroughly with deionized water and then dried in the oven at 40°C for about 12 h.

Deposition of Cu2O on TiO2 NAFs was conducted in a 50 mL Teflon-lined autoclave. The autoclave was filled with a solution containing 30 mL of EG and 10 mL of DMF, where a certain amount of CuSO4 (ranging from 0.005 to 0.05 g) was previously dissolved. The as-prepared TiO2 NAFs were immersed in the organic solution, and then the sealed autoclave was kept in an oven at 120°C for 16 h. After the autoclave was cooled down to room temperature naturally, the resulting samples were removed from the organic solution, washed several times with deionized water, and subsequently dried in an oven at 40°C for about 12 h.

2.2. Characterization and Photocatalytic Activity Evaluation of the Films

The surface morphology of the films was examined using a scanning electron microscope (SEM, Hitachi S-4700) operating at 15 kV. X-ray diffraction (XRD) analysis was performed on a Thermo ARL XTRA X-ray diffractometer using Cu Kα X-ray source. The chemical composition of the as-prepared film was characterized by an energy-dispersive X-ray spectrometer (EDS) attached to SEM operating at 15 kV. The ultraviolet-visible (UV-vis) diffuse reflectance spectra were recorded on a UV-2550 (SHIMADSU) spectrophotometer with BaSO4 as the reference. The photoelectrochemical property of the film electrodes was evaluated in a three-electrode cell using a Pt wire as counter electrode and a saturated calomel electrode (SCE) as reference electrodes. If needed, the working electrode could be irradiated from the front side by a sunlight-simulation lamp (Osram Ultra Vitalux 300W). The current with or without irradiation was measured in a 0.25 M Na2SO4 aqueous solution using a potentiostat (CHI 620B, CHI Co.). Photocatalytic activities of the samples were evaluated by the photodegradation of methyl orange (MO) solution with an initial concentration of 5.0 mg/L under simulated sunlight. The photodegradation experiments were conducted in a quartz reactor. In each test, one piece of the sample was hung in the liquid. Prior to irradiation, the suspension was kept in the dark for 60 min to achieve the adsorption-desorption equilibrium between the photocatalyst and methyl orange. Then, the solution was exposed to the light irradiation, and samples were taken at given time interval to analyze the concentration of MO by measuring the absorbance with the spectrophotometer.

3. Results and Discussion

To decorate TiO2 NAFs with nanosized Cu2O, a suitable condition for reduction of Cu2+ in organic solvent should be chosen. Figure 1(a) shows the XRD patterns of the products obtained by reduction of Cu2+ (2 mmol) in a mixture of EG and DMF at different temperatures. At a temperature of 120°C, all the diffraction peaks appearing in the XRD pattern of the product (indicated by the solid circles) can be indexed to cubic Cu2O phase (JCPDS number 65-3288), at which the peaks at 2θ values of 29.6°, 36.5°, 42.4°, and 61.5° correspond to 110, 111, 200, and 220 lattice planes of Cu2O, respectively. The broaden peaks indicate that the size of Cu2O is very small. The average crystal size calculated by Scherrer’s equation for (111) reflections of Cu2O is about 10 nm. The SEM image shown in Figure 1(b) reveals that these small-sized Cu2O nanocrystals are severely aggregated as a result of reduction in surface energy. When the reduction is conducted at 140°C, we can observe two new peaks at 2θ of 43.2° and 50.4° (indicated by open circles), which can be, respectively, assigned to the diffraction of (111) and (200) planes of cubic Cu (JCPDS number 04-0836). The result indicates the formation of Cu at 140°C. As the temperature is raised to 160°C, all the diffraction peaks of Cu2O disappear. Moreover, the peaks assigned to Cu become very sharp, suggesting that the growth of Cu crystals of large size occurs. This is confirmed by the SEM image of this sample (see Figure 1(c)), where Cu microcrystals can be observed. Therefore, we chose a temperature of 120°C to deposit Cu2O on TiO2 nanotube arrays, where the concentration of Cu2+ in the mixture of EG and DMF is changed to control the amount of Cu2O deposited on the films.

Figure 2(a) presents the digital photos of TiO2 NAFsbefore and after being treated in the mixture of EG and DMF containing different amounts of CuSO4 at 120°C for 16 h. Compared with the untreated TiO2 film, the treatment in CuSO4-contianing organic solution can result in an obvious color change, and the color of the treated TiO2 NAFs changes from light yellow to dull red as the amount of CuSO4 in the organic solution increases. These results hint that the deposition of Cu2O on the TiO2 NAFs occurs after treatment in CuSO4-contianing organic solutions, especially when the amount of CuSO4 is high. The XRD patterns of these untreated and treated TiO2 NAFs films are shown in Figure 2(b). In the XRD pattern of the untreated TiO2 film, the peaks at 35.1°, 38.4°, 40.2°, 53.0°, and 63.0° can be attributed to the background of Ti (JCPDS 44-1294). These peaks correspond to the diffraction of (100), (002), (101), (102), and (110) lattice planes of Ti, respectively. No diffraction peak attributed to TiO2 phase can be observed, indicating that the TiO2 nanotube arrays are amorphous. After being treated in CuSO4-contianing organic solutions at 120°C for 16 h, the TiO2 nanotube arrays are still in the amorphous state. In addition, in the XRD patterns of all treated NAFs we cannot observe any diffraction peak which can be assigned to the Cu2O phase, indicating that Cu2O might be well-dispersed on TiO2 NAFs.

The chemical composition of TiO2 NAFs before and after being treated in CuSO4-containing organic solution was characterized by EDS. Figure 3(a) shows the EDS spectra of these films, and the amount of Cu calculated from the spectra is presented in Figure 3(b) (see open circles). It is obvious that a small amount of fluorine is incorporated into TiO2 nanotubes, and treatment in CuSO4-containing organic solution cannot remove fluorine from the TiO2 nanotube layer. The observation of fluorine in the nanotube layer should result from the fact that F will migrate towards the Ti anode during the anodizing process [24]. From Figure 3(b), we can also find that the amount of Cu in the film shows an increasing trend as the concentration of CuSO4 in the organic solution increases. The increasing trend can be roughly fitted by a logarithm equation of . The surface morphology of treated TiO2 NAFs can be found in Figure 4. As can be seen from Figures 4(a) and 4(b), after being treated in organic solution containing a small amount of CuSO4 (viz. 0.001 g and 0.005 g), the amount of Cu2O decorated on the film (indicated by white arrows) is very small. When the amount of CuSO4 is raised to 0.01 g or 0.05 g (see Figures 4(c) or 4(d)), we can observe many nanoparticles deposited on the films (indicated by white arrows). These nanosized Cu2O particles are well dispersed on the wall of TiO2 nanotubes without blocking the nanotube, a commonly observed phenomenon in the case of deposition of Cu2O via electrochemical method [1619]. The observed result might be related to the fact that in the case of electrodeposition the applied voltage favors the formation of Cu2O crystals of large size [1619], which may easily block the nanotube, while in our case the growth of Cu2O resulting from the reduction of Cu(II) in organic solvent tends to form Cu2O nanocrystals (see Figure 1(a)). In addition, when Cu2O heterogeneously deposits on TiO2 NAFs, the interaction between Cu2O and TiO2 can reduce the surface energy of Cu2O nanocrystals and thus reduces their aggregation degree (namely, very large particles as observed in Figure 1(b) are not formed by aggregation of Cu2O nanocrystals). As a result, the nanotubes are not easily blocked by these small-sized Cu2O particles. Despite the observation of Cu2O nanoparticles on the film surface, however, these Cu2O nanoparticles cannot be detected by XRD (see Figure 2(b)). The possible reason for this is that these Cu2O nanoparticles are distributed on the film surface. Compared with the corresponding powder materials, nanosized materials dispersed on the film surface is often more difficult to be detected by conventional XRD technique when their crystal size or amount is not large enough. In addition, since the small-sized Cu2O particles tend to aggregate due to high surface energy (see Figure 1(b)), the particles of large size observed in Figure 4 might also be composed of several small-sized particles, which further make it difficult to detect Cu2O phase by XRD.

To determine whether the treatment of TiO2 NAFs in CuSO4-containing organic solution can extend their light-response scope from ultraviolet to visible light region, we have measured the UV-vis diffuse reflectance spectrum of the TiO2 NAFs treated in organic solution containing relatively high amount of CuSO4 (viz. 0.01 g and 0.05 g). Figure 5(a) shows the UV-vis diffuse reflectance spectrum of these two treated TiO2 NAFs. For comparison, the UV-vis diffuse reflectance spectrum of the untreated film is also shown. Compared with the untreated film, both two treated TiO2 NAFs exhibit significant increases in photoadsorption at the wavelength larger than 400 nm, suggesting that they can respond well to visible light. The absorption coefficient α follows the equation , where , , , , and are, respectively, plank constant, light frequency, proportionality coefficient that depends on the properties of the material, band gap, and a constant that can take different values depending on the type of electronic transition [25]. For a permitted direct transition, . Figure 5(b) shows the plot of against for three films, where the value of is obtained by extrapolating the linear part of the graphics to the axis of the abscissa (see dashed red lines). The band gaps estimated from the plots of verses photon energy ( ) are about 3.27, 2.20, and 2.16 eV for untreated TiO2 NAF and two treated TiO2 NAFs, respectively, (see Figure 5(b)). The observed decrease in the band gap after treatment in CuSO4-containing organic solutions is in line with the UV-vis adsorption spectra with a red shift. Since the band gap of Cu2O is about 2.2 eV, the value of 2.20 and 2.16 eV obtained for two treated NAFs confirms that Cu2O is deposited on TiO2 NAF.

The photoelectrochemical property of the as-prepared or Cu2O-decorated TiO2 film electrodes is investigated by measuring the anodic photocurrent in a 0.25 M Na2SO4 aqueous solution. As can be seen from Figure 6(a), without light irradiation (from 0–30 s), the dark currents for all films are almost equal to zero. When the light is on (from 30 to 60 s), the photocurrent increases sharply till reaching a certain value. If the light is off (from 60 to 90 s), the photocurrent declines rapidly to about zero. A similar phenomenon can also be observed in the range of 90–150 s. The observed photocurrent represents the anodic oxidation of water to oxygen by the photogenerated holes at the film electrode under light irradiation. For the untreated TiO2 film, under light irradiation the electrons are excited from the valence band to conduction band of TiO2 to form photogenerated electron-hole pairs. The photogenerated electrons and holes are separated under the external potential bias, and most electrons are transferred to titanium substrate to produce photocurrent with the hole oxidizing water to oxygen on the surface of the anode. It is clear from Figure 6(a) that the decoration of Cu2O can lead to a great rise in photocurrent density, where the photocurrent density for the treated TiO2 NAF in organic solution containing 0.05 g CuSO4 is about 3 times higher than that for the untreated TiO2 NAF. It is also interesting to note that the photocurrent density almost increases linearly with the amount of Cu in the film (see Figure 6(b)). These results suggest that the decoration of Cu2O can greatly improve the water splitting performance of the TiO2 NAFs under sunlight. The comparison of photocatalytic activity between untreated and treated TiO2 NAFs in organic solution containing 0.05 g CuSO4 is also evaluated by the photodegradation of MO. The MO aqueous solution shows an intense absorption band centered at ~464 nm and the peak intensity is proportional to its concentration. Figure 7 shows a comparison of the temporal evolution of the adsorption spectra of MO solution degraded by two films. It is obvious that the Cu2O-decorated TiO2 NAF exhibits better photocatalytic activity than the undecorated one. The degradation efficiency of MO for the Cu2O-decorated film reaches ~54.7% in 3 h, while that for the undecorated film is ~31.2%.

The enhanced activity of the Cu2O-decorated NAFs observed in our experiments can be attributed to the combined effect of several factors. Firstly, under simulated sunlight, TiO2 can be excited by UV light, and Cu2O can be excited by visible light, which will generate more electrons and holes for photocatalytic reactions as compared to undecorated TiO2 NAF. Secondly, the combination of TiO2 with Cu2O will lead to a reduced recombination of the photoexcited electrons and holes due to the difference between the band edges of Cu2O and TiO2 semiconductors. As shown in Figure 8, the electron excited under visible light may transfer from the conduction band of Cu2O to that of TiO2 since the conduction band edge for Cu2O is higher than that of TiO2 [23]. As a result, the recombination probability of the photoexcited electrons and holes will be reduced, leading to an improvement in photocatalytic activity.

4. Conclusions

In summary, we have presented a novel method for modification of anodic TiO2 nanotube array films with Cu2O nanoparticles. The method is based on the theory of heterogeneous nucleation and growth in an organic solvent (ethylene glycol and N,N-dimethylformamide) containing CuSO4. The Cu2O nanoparticles are found to be well dispersed on the wall of TiO2 nanotubes without blocking the nanotube, and the amount of Cu2O deposited on the TiO2 nanotube array films shows an increasing trend as the concentration of CuSO4 increases. The decorated nanotube array films can respond well to both ultraviolet and visible light and show much better photocatalytic activity than the undecorated film.

Acknowledgment

This work was supported by the Natural Science Foundation of Zhejiang Province (no. LY12B07011).