The development of visible-light-responding photocatalysts has become one of the most challenging and urgent topics that needs addressing because of the increasingly grim environmental and energy issues that confront mankind today [1-3]. Since Fujishima et al. [4] reported on water photolysis with titanium dioxide (TiO2) as the anode in 1972, TiO2 has been deemed as the most promising photocatalyst due to its outstanding performance, low cost, nontoxicity and stability. However, the shortcomings of a wide bandgap (3-3.2 eV) and fast photogenerated carriers recombination for TiO2 hinder its extensive application in photocatalysis [3, 5]. Therefore, improving its visible-light utilization and suppressing the photogenerated carriers recombination are significant foci in the field of photocatalysis.
The sensitization of TiO2 by narrow bandgap semiconductors has proven to be an effective strategy to enhance its visible-light photocatalytic performance [3, 6], resulting in the transfer of the photogenerated carriers from the narrow bandgap semiconductors to TiO2. In the past, cuprous oxide (Cu2O), a fascinating p-type semiconductor with a 2.0 eV bandgap, has received significant research interest as a visible-light-responding photocatalyst [7-12]. However, the visible-light photocatalytic activity of bare Cu2O is commonly comparatively low due to the rapid recombination of photogenerated carriers [13]. In the past few decades, increasing interest has been aroused on the combination of TiO2 with Cu2O to form a visible-light-responding heterostructure to be used as photocatalysts and photovoltaic cells because of their favorably matched band structures [14-25].
Indium orthovanadate (InVO4), an n-type semiconductor with a bandgap of 2.0 eV, has also enjoyed considerable attention as a visible-light-responding photocatalyst for hydrogen evolution by water splitting [26-28] and organic pollutants degradation [29, 30]. Similarly, InVO4 commonly exhibits negligible photocatalytic activity because of its poor adsorption capability and facile photogenerated carrier recombination [31]. Therefore, much research effort has been directed towards the development of TiO2-InVO4 composite photocatalyst [31-39].
Based on the strategy of energy gap engineering, compositing p-type Cu2O and n-type InVO4 semiconductors with n-type semiconductor TiO2 to construct a novel ternary heterostructure can be expected to display improved visible-light photocatalytic performance because of coupling of their different energy levels and the unique properties of the formed heterostructures. Herein, we demonstrate a facile wet chemistry route to fabricate a novel InVO4-Cu2O-TiO2 ternary nanoheterostructure that shows enhanced photocatalytic performance compared with the InVO4-TiO2 and Cu2O-TiO2 binary nanoheterostructures used for the degradation of methyl orange (MO) under visible-light irradiation. Furthermore, for the purpose of practical application, an ordinary 9 W energy-saving fluorescent lamp was used as the visible-light resource. To the best of our knowledge, no works have been documented on the fabrication of InVO4-Cu2O-TiO2 ternary nanoheterostructure for organic pollutant photodegradation.
All chemical reagents used in the experiments were of analytical purity and employed without further purification.
InVO4 nanoparticles were prepared by a facile hydrothermal method similar to that reported by Ge et al [32]. In a typical experiment, 0.001 mol InCl3 was dissolved in 10 mL deionized water, and 20 mL of 0.05 mol/L NH4VO3 solution was then added dropwise under magnetic stirring. The pH value of the resulting mixture was adjusted to about 7 with 2 mol/L NaOH solution. After that, the mixture was kept stirring for 30 min and sonicating for 10 min before being transferred into a 60 mL Teflon-lined stainless steel autoclave. Hydrothermal reaction then proceeded at 120 ℃ for 8 h. Finally, the produced white slurry was centrifuged, washed with deionized water 3 times and dried at 60 ℃ for 12 h to give white InVO4 nanoparticles.
An InVO4-TiO2 nanoheterostructure was synthesized by a sol-gel method. Typically, 0.0667 g of the as-prepared InVO4 was dispersed in a mixture solvent of 5 mL deionized water and 5 mL ethanol, and 10 mL of 0.1% cetyltrimethylammonium bromide (CTAB) solution was then added and followed by sonicating for 10 min. A solution composed of 1.42 mL tetrabutyl titanate and 18.58 mL anhydrous alcohol was also added under magnetic stirring, and the resulting suspension was kept stirring for 30 min. Finally, the produced white slurry was centrifuged, washed with deionized water 3 times, dried at 60 ℃ for 12 h and calcined at 450 ℃ for 3 h to give the 16.67%InVO4- 83.33%TiO2 nanoheterostructure. In addition, TiO2, 5%InVO4- 95%TiO2 and 10%InVO4-90%TiO2 were prepared via a similar route by changing the starting material ratios.
A Cu2O-TiO2 composite was synthesized via an easy precipitation route similar to that reported by Huang et al. [15]. 0.4159 g Cu(Ac)2·H2O was dissolved in 50 mL anhydrous ethanol to obtain a deep green solution, and 0.15 g of the as-prepared TiO2 was dispersed in the solution assisted by sonication for 10 min. 50 mL glucose solution (0.13 mol/L) as reducing agent and 60 mL NaOH solution (0.3 mol/L) in the mixed solvent of 35 mL anhydrous ethanol and 25 mL deionized water were added dropwise to the suspension in sequence under magnetic stirring, and then the resulting suspension was heated to 80 ℃ and kept stirring for 30 min. After the suspension was cooled to room temperature, the precipitates were collected by centrifugation, washed with anhydrous ethanol 2 times and deionized water 2 times, and then dried at 60 ℃ for 12 h in vacuum. Cu2O, 30%Cu2O-70%TiO2 and 70%Cu2O- 30%TiO2 were also prepared by a similar route through altering the starting material ratios.
The as-prepared InVO4-TiO2 nanoheterostructure was coupled with Cu2O by a simple precipitation route. In a typical experiment, 0.3328 g Cu(Ac)2·H2O was dissolved in 50 mL anhydrous ethanol, and 0.18 g of the as-prepared 16.67%InVO4- 83.33%TiO2 nanoheterostructure was dispersed in the solution followed by sonication for 10 min. 50 mL glucose solution (0.1 mol/L) and 60 mL NaOH solution (0.3 moL/L) in the mixed solvent of 35 mL anhydrous ethanol and 25 mL deionized water were added dropwise to the suspension in sequence under magnetic stirring, and the resulting suspension was then heated to 80 ℃ and stirred constantly for 30 min. After the suspension was cooled to room temperature, the precipitates were collected by centrifugation, washed with anhydrous ethanol 2 times and deionized water 2 times, and dried at 60 ℃ for 12 h in vacuum. Additionally, 5%InVO4-40%Cu2O-55%TiO2, 20%InVO4-40%Cu2O-40%TiO2 and 30%InVO4-40%Cu2O- 30%TiO2 were fabricated via similar routes by changes to the starting material ratios.
Phase structures of the samples were determined by X-ray powder diffraction (XRD) taken on a Rigaku-D-Max rA 12 kW diffractometer with Cu Kα radiation (λ = 1.54056 ) at 40 kV and 300 mA. A JEM-2010 transmission electron microscope (TEM) and a JEOL-3010 high-resolution transmission electron microscope (HRTEM), operated at an acceleration voltage of 200 kV, were employed to observe the morphologies and sizes of the samples. X-ray photoelectron spectra (XPS) of the samples were collected on a K-Alpha 1063 XPS system with Al Kα radiation. Diffuse reflectance ultraviolet-visible spectra (UV-vis/DRS) of the samples were recorded on a Beijing Purkinje TU-1901 UV-vis spectrophotometer equipped with a diffuse reflectance accessory with an IS19-1 integrating sphere and BaSO4 powders as reference.
The visible-light photocatalytic activities of the samples were investigated in a self-assembly reactor using photocatalytic degradation of MO as the probe reaction and an ordinary 9 W energy-savingfluorescent lamp as the visible-light irradiation source. 0.15 g photocatalyst was put into 200 mL MO solution (20 mg/L). Prior to irradiating, the suspension was magnetically stirred in the dark for 30 min to establish an adsorption/desorption equilibrium, and then exposed to the lamp with a distance of 10 cm. Samples were taken from the reaction suspension at 60 min intervals. The upper, lucid liquid obtained after centrifugal separation was analyzed by a UV-vis spectrophotometer at the maximum absorption wavelength of MO of 464 nm. The total organic carbon (TOC) removal of MO solution was monitored on a Shimadzu TOC-VCPH analyzer.
The XRD patterns of the as-prepared InVO4, Cu2O, TiO2, 70%Cu2O-30%TiO2, 10%InVO4-90%TiO2, 10%InVO4- 40%Cu2O-50%TiO2, together with the standard patterns of orthorhombic InVO4 (PDF 48-0898), anatase TiO2 (PDF 21-1272) and cubic Cu2O (PDF 05-0667) are shown in Fig. 1. All the diffraction peaks of individual InVO4, Cu2O and TiO2 in Fig. 1 match well with the standard patterns of orthorhombic InVO4, cubic Cu2O and anatase TiO2, respectively. These three phases can be readily found in the XRD patterns of their corresponding composites. However, it should be noted that only the strongest (001) crystal plane peak at ca. 2θ = 32.9° for the InVO4 phase could be detected in 10%InVO4-40%Cu2O-50%TiO2. This may be due to the lesser amount and high dispersion of InVO4 in 10%InVO4-40%Cu2O-50%TiO2 relative to the other systems studied. This agrees well with Min et al’s report [37] in which no XRD diffraction peaks of InVO4 appeared in an InVO4-TiO2 composite with a mass ratio of 1:10. In addition, no other clear peaks for impurities can be detected in Fig. 1, suggesting high purity in these compounds and composites.
From the TEM images of the 10%InVO4-40%Cu2O-50%TiO2 nanoheterostructure (Fig. 2(a) and (b)), it can be clearly observed that the obtained nanoparticles are in the size of 5-20 nm in diameter with agglomeration, which is exactly what is needed for the purpose of forming heterostructures resulting from contact of the nanosized TiO2, Cu2O and InVO4. Furthermore, we can observe, from the corresponding HRTEM images, the formation of nanoheterostructures, as shown in Fig. 2(c) and (d). In addition, the fringe spacing values, measured to be 0.36, 0.24 and 0.27 nm, can be ascribed to the d-spacing values for (101) lattice planes of anatase TiO2, (111) lattice planes of cubic Cu2O and (112) lattice planes of orthorhombic InVO4.
The photocatalytic activities of the as-prepared InVO4, Cu2O, TiO2, InVO4-TiO2, Cu2O-TiO2 and InVO4-Cu2O-TiO2 with different mass ratios for the MO degradations are illustrated in Fig. 3. From Fig. 3(a), it can be seen that bare TiO2 and InVO4 exhibit no discernable photocatalytic activities, and the highest MO degradation rate is only 8% for the 10%InVO4-90%TiO2 nanoheterostructure, obtained after illumination with the 9 W energy-saving fluorescent lamp for 5 h. As shown in Fig. 3(b), the bare Cu2O also shows no photocatalytic activity. However, when coupling Cu2O with TiO2, the resulting nanoheterostructures reveal considerable activities. The MO degradation rate for the 70%Cu2O-30%TiO2 nanoheterostructure achieves 84% after illumination with the 9 W energy-saving fluorescent lamp for 5 h. Meanwhile, it can be also found that the MO degradation rate in the Cu2O-TiO2 nanoheterostructures increases upon an increase in the amount of Cu2O. As for the TiO2-Cu2O-InVO4 ternary nanoheterostructures, their initialphotocatalytic activities are greater than those for the TiO2-Cu2O binary nanoheterostructures. For example, the MO degradation rate of 5%InVO4-40%Cu2O-55%TiO2 reaches 62.9% after illumination for 2 h, as displayed in Fig. 3(c), while it is 46.7% for 70%Cu2O-30%TiO2. By prolonging the illumination time to 5 h, the MO degradation rate of 5%InVO4-40%Cu2O-55%TiO2 is 86.3%, which is slightly higher than that for 70%Cu2O- 30%TiO2. However, it should be pointed out that the as-mentioned MO degradation rates of 5%InVO4-40%Cu2O- 55%TiO2 were not the maxima in our experiments, as depicted in Fig. 3(c). After illumination for 2 h, the MO degradation rate of 30%InVO4-40%Cu2O-30%TiO2 attains 65.4%, and the MO degradation rate of 10%InVO4-40%Cu2O-50%TiO2 reaches 89.13% after illumination for 5 h, which are considerably greater than those of 70%Cu2O-30%TiO2.
The mineralization of MO (i.e. TOC removal) under visible-light irradiation for 10%InVO4-40%Cu2O-50%TiO2 is presented in Fig. 4. 10%InVO4-40%Cu2O-50%TiO2 affords only 32.17% TOC removal after illumination for 5 h, which is a common phenomenon for the MO photocatalytic degradation [40-42]. This may be due to the fact that MO degrades to stable intermediates such as aniline, phenol, benzene, benzene sulfonic acid, aliphatic acid or aldehyde [40, 42], and therefore requires more time to achieve greater TOC removal.
To investigate the stabilities of the 50%Cu2O-50%TiO2 and 10%InVO4-40%Cu2O-50%TiO2 nanoheterostructures for the MO photocatalytic degradations under illumination with the 9 W energy-saving fluorescent lamp, we repeated the photocatalytic degradation experiments employing the same photocatalysts six times. The results are shown in Fig. 5. After six cycles, the MO degradation rate of 10%InVO4-40%Cu2O-50%TiO2 maintains at ca. 80% and the degradation efficiency is maintained at over 90%, which is exactly the requirement needed for application to organic pollutant photocatalytic degradation. However, the MO degradation rate of 50%Cu2O-50%TiO2 is around 74%. We also find that the initial photocatalytic activity of 10%InVO4-40%Cu2O-50%TiO2 in the first cycle is higher than that of 50%Cu2O-50%TiO2, which is in good agreement with the case described for Fig. 3.
Generally, Cu2O is an efficient photocatalyst, but is not stable [43]. To examine the stability of Cu2O, XPS characterization was performed for the used 10%InVO4-40%Cu2O-50%TiO2, and the corresponding high-resolution XPS spectrum for the surface Cu 2p is displayed in Fig. 6. The peaks centered at binding energies of 932.5 and 952.6 eV can be attributed to the Cu 2p3/2 and Cu 2p1/2 peaks, respectively, confirming the existence of Cu(Ⅰ) species [44]. Meanwhile, a shakeup satellite peak (denoted as Sat) can be found at 944.2 eV, suggesting the appearance of Cu(Ⅱ) species [44]. Due to the fact that the contents of compositions are commonly proportional to the peak areas [45], the small peak area of the Sat peak for the Cu(Ⅱ) species suggests that a small fraction of Cu2O was oxidized to CuO during the use of 10%InVO4-40%Cu2O-50%TiO2, thus implying bearable photocorrosion of Cu2O.
The UV-vis/DRS spectra of TiO2, InVO4, Cu2O, 10%InVO4- 90%TiO2, 50%Cu2O-50%TiO2 and 10%InVO4-40%Cu2O- 50%TiO2 are presented in Fig. 7(a). The bare TiO2 can only absorb UV light with a wavelength shorter than 400 nm. The bare InVO4 has a weak absorbance in the wavelength range of 400-650 nm. Therefore, 10%InVO4-90%TiO2 also shows a weak absorbance in this wavelength range. Cu2O exhibits the strongest absorbance with the wavelength extended to greater than 800 nm. 50%Cu2O-50%TiO2 and 10%InVO4-40%Cu2O- 50%TiO2 also display strong absorbances in the visible light region with the wavelength greater than 800 nm. Although Cu2O and 50%Cu2O-50%TiO2 reveal wider and stronger visible light absorption than 10%InVO4-40%Cu2O-50%TiO2, this does not mean that the photocatalytic activities of Cu2O and Cu2O-TiO2 under visible light are greater than that of InVO4-Cu2O-TiO2. This is because there exists an implicit conflict between wide-range visible light absorption and adequate redox capability, and a high quantum yield cannot be achieved simply by extending the absorption of visible light [3]. On the other hand, it is well known that the recombination rate of the photogenerated carriers for a photocatalyst plays a vital role in its photocatalytic activity.
To estimate the bandgap energies of the above photocatalysts, plots of (ɑhʋ)1/2 versus hʋ are presented in Fig. 7(b), where ɑ and hʋ are the absorption coefficient and discrete photon energy, respectively. The bandgap energies of photocatalysts can be obtained from the extrapolated values of the tangents to the X-axis. Therefore, the bandgap energies of bare TiO2, InVO4 and Cu2O are estimated to be 3.13, 2.11 and 2.02 eV, respectively, which are very close to the reported values [15, 39]. The bandgap energies of 10%InVO4-90%TiO2, 50%Cu2O-50%TiO2 and 10%InVO4-40%Cu2O-50%TiO2 are estimated as 3.17, 2.58 and 2.78 eV, respectively, suggesting that the introduction of Cu2O can efficaciously decrease the bandgap energies of composites. The high bandgap energy for 10%InVO4-90%TiO2 may account for its extremely low visible-light photocatalytic activity, as shown in Fig. 3(a).
Fig. 8 illustrates the bandgap configurations of Cu2O, InVO4 and TiO2 [15, 39] and the charge separation and transfer behaviors between them under visible-light irradiation. As indicated in Fig. 8, both Cu2O and InVO4 can be excited under visible-light irradiation to generate electron-hole pairs due to their narrow bandgaps. The conduction and valence band edge potentials of Cu2O (−1.6 and +0.4 eV vs. SHE, respectively) and InVO4 (−0.5 and +1.5 eV vs. SHE, respectively) are more negative than those of TiO2 (−0.23 and +2.97 eV vs. SHE, respectively). As a result, the photogenerated electrons on the Cu2O and InVO4 conduction bands can transfer to the TiO2 conduction band. On the other hand, the conduction and valence band edge potentials of Cu2O are also more negative than those of InVO4, leading to the photogenerated electron injection from the Cu2O conduction band to the InVO4 conduction band and the photogenerated hole migration from the InVO4 valence band to the Cu2O valence band. Therefore, the spatial separation of photogenerated electron-hole pairs can effectively promote the visible-light photocatalytic performance of the TiO2-Cu2O-InVO4 ternary nanoheterostructure. Moreover, two p-n semiconductor heterojunctions can be formed by p-type Cu2O with respective n-type TiO2 and InVO4 upon their close contact, as observed in Fig. 2(c) and (d). The resulting inner electric fields with orientations from TiO2 and InVO4 to Cu2O can further boost the migration of photogenerated electron-hole pairs motivated by the potential energy difference.
A novel visible-light-responding TiO2-Cu2O-InVO4 ternary nanoheterostructure with excellent photocatalytic performance relative to TiO2, Cu2O, InVO4, Cu2O-TiO2 and InVO4-TiO2 has been successfully fabricated by facile wet chemistry methods. With 10%InVO4-40%Cu2O-50%TiO2 as the photocatalyst, the MO degradation rate achieves close to 90% under illumination by a 9 W energy-saving fluorescent lamp for 5 h, and the degradation efficiency is maintained at over 90% after six cycles. This can be ascribed to the band configuration optimization of TiO2, Cu2O and InVO4 in the composite and the formation of two p-n semiconductor heterojunctions by p-type Cu2O with n-type TiO2 and InVO4. The excellent performance of TiO2-Cu2O-InVO4 fits the requirements needed for its application to photocatalytic degradation of organic pollutants.
2016,
Vol. 37 Issue (6): 855-862 
PDF (21874 KB)
2016,
Vol. 37 Issue (6): 855-862 PDF (21874 KB)