photocatalyst was successfully prepared via solid state reaction and further combined with TiO2 by a sol-gel method and a solid phase method, respectively. For comparison, Pt particles were loaded to prepare Pt- via a hydrogen reduction method. All the samples were characterized by thermogravimetry and differential thermal analysis (TG/DTA), X-ray diffraction (XRD), UV-vis diffuse reflectance spectroscopy (UV-vis DRS), and scanning electron microscopy (SEM) techniques. Photocatalytic H2 evolution activities of the as-obtained samples were evaluated from aqueous oxalic acid solution under simulated sunlight irradiation. The effects of photocatalyst concentration, TiO2 content, and composite method on the H2 evolution activities of the as-obtained photocatalysts were investigated. The results show that, when the concentration of photocatalyst is 0.8 , the TiO2- composite photocatalyst prepared by a sol-gel method exhibits the optimized photocatalytic activity, and the H2 production rate is 4.35 mmol  with 30 wt.% content of TiO2.

1. Introduction

Water splitting with photon energy has been extensively studied since the Honda-Fujishima effect was reported [1, 2]. TiO2-based photocatalytic reactions such as organic matter degradation [3, 4], H2 evolution [5], and disinfection [6] have received great attention around the world. Conventional TiO2 photocatalysts possess excellent activity and stability, but require near-ultraviolet (UV) irradiation (about 4% of the solar energy) for the effective photocatalysis, which severely limits their practical applications [710]. One of the promising methods to solve the above problem for TiO2 photocatalysts is to modify TiO2 with metals or nonmetals [1113]. However, although the doping leads to significant visible light absorption for TiO2, the role of dopant metals or nonmetals on the photocatalytic activity is still under investigation. Indeed, the charges recombination processes largely depend on both dopant concentration and type. Low concentration is not sufficient to significantly increase the absorption of the visible light in TiO2. High dopant concentration causes an efficient narrowing of the TiO2 band gap but at the same time it probably introduces new recombination centers for the photo-generated charges. From these consideration, a higher efficiency photocatalyst is needed to be developed. Another method is to design efficient non-TiO2 visible-light activated photocatalysts. These novel kinds of photocatalysts, such as K2Ti4O9 [14, 15], K2La2Ti3O10 [16], K4Nb6O17 [17, 18], and KLaNb2O7 [19], show potentially applications in the decomposition of pure water due to their unique structures such as layered or tunnel structures [20]. However, these photocatalysts, in general, display small absorption in the visible light region. Y2Cu2O5 is a good p-type semiconductor with a special layered structure and exhibit good electromagnetic properties [21, 22]. However, there is comparatively little attentions given to photocatalytic activity of Y2Cu2O5. Due to its unique interlayer, where the electron-hole recombination process could be retarded by physical separation of the electron-hole pairs generated by photo-absorption. We try to prepare the novel Y2Cu2O5-based photocatalysts for water splitting. Furthermore, the photocatalysts are modified by noble metal Pt and TiO2 nanoparticles to enhance the photocatalytic activity.

2. Experimental

2.1. Synthesis of Photocatalyst
2.1.1. Preparation of Y2Cu2O5 Microparticles

All the chemical reagents were analytically pure and used without further purification. The photocatalyst Y2Cu2O5 was prepared by a solid phase method. CuO and Y2O3 powders were weighed according to the stoichiometric ratio, mixed together and grinded fully in an agate mortar, then heated for 12 h in air, and slowly cooled to room temperature. After milling and mixing, the powders were calcined at a certain temperature for 12 h three times in air. Y2Cu2O5 microparticles were then obtained.

2.1.2. Preparation of Pt-Y2Cu2O5 Microparticles

Y2Cu2O5 (3 g) was added into a 1 mM H2PtCl6 aqueous solution, and the total volume of the mixture was 100 mL by adding water. After ultrasonic treatment for 30 minutes, the mixture was stirred at 80°C with H2 passing through (flow rate of 330–340 mL min−1) for 2 h to reduce Pt4+. After being filtered, the precipitates were washed by deionized water and dried at 100°C to obtain Pt-Y2Cu2O5 microparticles.

2.1.3. Preparation of TiO2-Y2Cu2O5 Composite Microparticles

The photocatalyst TiO2-Y2Cu2O5 was prepared via two routes.(i)Sol-gel approach: 10 mL Ti(OBu)4 and 20 mL ethanol were vigorously stirred to get pale yellow transparent solution (A) at room temperature. 1.4 mL HNO3 was slowly dropped into the mixture of 6 mL water and 20 mL ethanol under stirring to get the mixture (B). Solution (A) was dropwise added into the mixture (B) under vigorous stirring over 1 h, and TiO2-Y2Cu2O5 sol was prepared by using appropriate amount of Y2Cu2O5. The as-obtained gel was subsequently transferred into an oven and kept at 100°C for 1 h. The as-prepared precursor was then annealed at 400°C for 2 h. The crystalline phase of TiO2 is mainly due to the anatase phase. (ii)Solid state reaction: The as-prepared Y2Cu2O5 and P25 TiO2 with a certain ratio were added into 100 mL ethanol under stirring. After that, the precipitates from the mixture were collected by centrifugation, and then rinsed with ethanol several times. The as-prepared sample was dried in an oven at 80°C for 2 h. The final product was prepared via calcination at 400°C for 2 h.

2.2. Photocatalyst Characterization

Thermogravimetric and differential thermal (TG-DTA) patterns of the samples were recorded by using thermogravimetric-differential thermal analyzer to determine the temperature of possible decomposition and phase changes. The crystal structures of the as-prepared photocatalysts were identified by powder X-ray diffraction (XRD, Bruker D8) using Cu Kαradiation (=1.5418 Å) at a scan speed of 0.05° s−1, a voltage of 40 kV and a current of 300 mA. The surface morphologies of the samples were observed with a JEOL JSM-5600LV scanning electron microscope (SEM) operated at 25 kV. The UV-vis diffuse reflectance spectra (UV-vis DRS) of the as-prepared photocatalysts were measured on a UV-vis spectrometer (Shimadzu UV-2500). Brunauer-Emmett-Teller (BET) surface areas of the samples were determined by nitrogen adsorption using a Micromeritics ASAP 2000 system.

2.3. Photocatalytic Activity Measurement

The photocatalytic activities of the as-prepared photocatalysts were evaluated by monitoring the amount of hydrogen evolution from aqueous oxalic acid solution under simulated sunlight irradiation. Typically, the photocatalytic reactor consists of two parts, a quartz cell (600 mL) with a circulating water jack and a Xe lamp (150 W, wavelength of 290–800 nm (unfiltered), main wavelength of 400–700 nm) placed inside the quartz cell. In all experiments, the reaction temperature was kept at room temperature by using the circulating water jack to prevent any thermal effect. In all experiments, 100 mL of deionized water containing a specified amount of the as-prepared photocatalyst and 0.05 M H2C2O4 mixed sacrificial agent was added into the reactor. Subsequently, the reactor was left in dark and N2 was bubbled through the reaction mixture for 30 min to remove O2. Afterwards, the photocatalytic reaction system was closed and exposed to the light irradiation. During the irradiation, the mixture was suspended by using a magnetic stirrer within the quartz cell. The as-produced H2 was collected in a water manometer and analyzed by gas chromatography (TCD, N2 as gas carrier, zeolite NaX column). An eliminator containing saturated NaOH solution was placed before the collector of H2 to remove CO2. The volume of H2 was determined by the volume of the displaced water at different intervals of irradiation time. Some experiments were repeated three times and the results were reproducible within the experiments errors (±4%).

3. Results and Discussion

3.1. TG/DTA Curves

As seen in Figure 1, the weight and heat changes of samples are relatively uniform, and the weight loss is only 2% under 800°C; suggesting that the precursors CuO and Y2O3 have no mutual reaction under this temperature. However, there exists an obvious weight loss stage with the weight loss of 4% at about 800~1000°C, the corresponding small exothermic DTA is observed also, probably due to the transformation of CuO and Y2O3 into Y2Cu2O5 [23], and a certain amount of other compounds, such as YCuO2 (shown in Figure 2), which result in weight loss.

3.2. XRD Patterns

Figure 2 shows the XRD patterns of the samples calcined at 800°C, 850°C, 900°C, 950°C, and 1000°C, respectively. From Figure 2, most diffraction peaks from this pattern can be indexed to reflections of CuO and Y2O3 except for a minor peak ascribed to a small amount of Y2Cu2O5 at 800°C. This indicates that a solid-state reaction starts around 800°C, resulting in the formation of Y2Cu2O5. The results are in good agreement with the reported data [23]. With further increasing the calcination temperature, the peaks of both CuO and Y2O3 start to decrease in intensity and more Y2Cu2O5 is formed. While calcination temperature increase to 950°C and 1000°C, respectively, the diffraction peaks are mainly indexed to Y2Cu2O5 (JCPDS Card No. 84-1853), and a certain amount of other compound, such as YCuO2 (JCPDS Card No. 76-1422), indicating that most of CuO and Y2O3 are transferred to Y2Cu2O5. But the diffraction peaks of Y2O3 and CuO still exist, although at very small intensities.

3.3. Specific Surface Area

The specific surface areas of Y2Cu2O5 photocatalysts calcined at 800 and 950°C are listed in Table 1. With the increase of calcination temperatures, the specific areas increase from 1.781 to 2.151 m2 g−1, while the mean pore size and the pore volume of Y2Cu2O5 particles decrease. The photocatalytic activity relates to the particle size and the specific surface area for photocatalyst [24]. Although the increase of calcination temperature can cause particle size growth and aggregation among the particles, Y2Cu2O5 remains the major component in the sample calcined at 950°C, which was confirmed by XRD analysis (see Figure 2). That is, most of CuO and Y2O3 are transferred to Y2Cu2O5 at 950°C. Furthermore, in the initial synthesis stage, the as-produced Y2Cu2O5 crystallite size is also small.

3.4. UV-Vis DRS Spectra

The UV-vis DRS spectra of Y2Cu2O5, TiO2 and 1 wt.% Pt/Y2Cu2O5 are compared in Figure 3. The Y2Cu2O5 photocatalyst shows low reflectance compared to TiO2, revealing good absorption ability. The absorption edge of Y2Cu2O5 is located at 500 nm, and the energy band gap of Y2Cu2O5 can be calculated to be 2.48 eV according to the following formula [25]. The results show that the Y2Cu2O5 photocatalyst shows excellent response performance under visible light.

The reflectance of 1 wt.% Pt/Y2Cu2O5 decreases a little compared to Y2Cu2O5, indicating the superior absorption ability of 1 wt.% Pt/Y2Cu2O5 compared to Y2Cu2O5, which is favorable for the enhancement of the photocatalytic activity. The order of absorption abilities of the samples is as follows: TiO2 < Y2Cu2O5 < 1 wt.% Pt/Y2Cu2O5, which results from the contribution of metal nanoparticles plasma absorption peak and the formation of heterojunction called Schottky junction with the semiconductor Y2Cu2O5 loaded by a certain amount of Pt.

3.5. SEM Morphology

The SEM images of the as-prepared photocatalysts calcined at 950°C were shown in Figure 4. From Figure 4(a), the TiO2-Y2Cu2O5 photocatalyst prepared by the sol-gel approach is of small spherical shape with agglomeration. The surfaces of some large Y2Cu2O5 particles are covered by tiny TiO2 particles to form heterogeneous cladding structures. Furthermore, the rough surfaces of particles can offer a large amount of sites for the photocatalytic reaction. Meanwhile, the TiO2-Y2Cu2O5 particles prepared by the solid phase method are more irregular than that prepared by the sol-gel approach (see Figure 4(b)). The Y2Cu2O5 photocatalyst in Figure 4(c) is of irregular shape with less aggregation.

3.6. Photocatalytic H2 Evolution Activity
3.6.1. Effect of the Photocatalyst Concentration

The H2 evolution activities of Y2Cu2O5 photocatalysts calcined at 950°C with different concentrations are shown in Figure 5. The H2 evolution activity increases with the increase of the photocatalyst concentrations. However, the H2 evolution activity reaches its maximum at the photocatalyst concentration of 0.8 g L−1. The highest photocatalytic H2 evolution activity of 3.78 mmol h−1 g−1 is obtained, when the initial concentration of 0.05 mol L−1 of oxalic acid is used as the sacrificial reagent. It is well known that the active center number on catalyst particle surfaces and the penetration ability of incident light in reactor are extremely important for the photocatalytic H2 evolution activity [26]. Generally, larger catalyst concentration provides more catalytically active centers for the absorption of photons, and then more electrons and holes are generated. However, the excess photocatalyst may act as an optical filter and impede the further penetration of incident light into the suspension. We observed that the photocatalytic H2 evolution activity increases with increasing the Y2Cu2O5 concentrations when the concentration is lower than 0.8 g L−1, whereas it decreases with further increasing Y2Cu2O5.

3.6.2. Effect of Pt Loading Content on Photocatalytic Activity

The effect of Pt loading content on the H2 evolution activity of Pt-loaded Y2Cu2O5 is shown in Figure 6. After loading Pt, the photocatalytic activity is obviously improved. The H2 evolution rate increases from 3.78 mmol h−1 g−1 to 4.12 mmol h−1 g−1 at first and then decreases with increasing the Pt loading content and reaches the summit of 4.12 mmol h−1 g−1 when the Pt loading content is 1 wt.%, which increases by ca. 9% in comparison with that of the pure Y2Cu2O5 (3.78 mmol h−1 g−1). Noble-metal nanoparticles (NPs) such as Pt, Au, and Ag can respond to visible light due to the localized surface plasmon resonance (LSPR). LSPR is produced by the collective oscillations of the surface electrons, exhibiting great potential for extending the light absorption range of semiconductors [2729]. The enhancement of H2 evolution activity at low fraction of Pt deposition can be ascribed to the formation of the heterojunction called Schottky junction between metal Pt and semiconductor Y2Cu2O5, and the electrons and holes generated by light irradiation can be localized at Pt and Y2Cu2O5, respectively. It is supposed that the loaded noble metal particles could serve as efficient electron traps, which suppressed the recombination rate of photogenerated electron-hole pairs and therefore enhanced the photocatalytic activity [30]. It is clear that there exists an optimum loading amount for Pt particles. The appropriate amount of Pt particles loaded on sample surface can trap the larger number of photoexcited electrons, resulting in the enhanced photocatalytic activity. However, excess Pt particles would serve as the recombination centers that decrease the photocatalytic activity. Moreover, the excess metal particles can mask the Y2Cu2O5 surface and reduce the light absorption capability of the catalyst and therefore reduce the photoexcitation to generate the active electrons. In addition to the shadow effect (i.e., cover on the surfaces of Y2Cu2O5), there are a few other possible reasons that affect the hydrogen generation due to the excessive Pt. It can also affect the crystalline formation, particle boundary and agglomeration, and so forth. Further detailed studies are in progress to understand the mechanism.

3.6.3. Effect of TiO2 Content on Photocatalytic Activity

After Y2Cu2O5 is combined with TiO2, the formation of heterojunction at the interfaces of Y2Cu2O5 (p-type semiconductor) loaded with TiO2 (n-type semiconductor) can improve the separation of electrons and holes, sequentially, enhancing the adsorbed water splitting into H2. As shown in Figure 7, the H2 evolution activities of TiO2-Y2Cu2O5 prepared by a sol-gel method with different TiO2 contents are in the order of 30 wt.% TiO2-Y2Cu2O5 > 20 wt.% TiO2-Y2Cu2O5 > 40 wt.% TiO2-Y2Cu2O5 > 50 wt.% TiO2-Y2Cu2O5 > 70 wt.% TiO2-Y2Cu2O5. It can be concluded that the introduction of small amount of TiO2 onto the Y2Cu2O5 surfaces results in the effective heterojunction formation with forming good transfer channels to separate electrons and holes. The TiO2 content of 30 wt.% in Y2Cu2O5 provides the highest efficiency for photocatalytic H2 evolution. However, when the amount of TiO2 in Y2Cu2O5 increases to a certain extent, excess TiO2 cannot contribute to the heterojunction formation, it only displays its photocatalytic activity, which can only use about 3%–5% of UV light in sunlight for the H2 evolution.

3.6.4. Effect of Composite Method on Photocatalytic Activity

Y2Cu2O5 combined with TiO2 via two different methods of sol-gel and solid phase methods in the experiments. The effect of the two different composite methods on H2 evolution activities of photocatalysts is shown in Figure 8. The H2 evolution rate of TiO2-Y2Cu2O5 prepared by the sol-gel approach reaches 4.35 mmol h−1 g−1, which is higher than that of TiO2-Y2Cu2O5 prepared by the solid state method (3.99 mmol h−1 g−1). It may be due to the fact that in the sol-gel process, TiO2 can combine sufficiently with Y2Cu2O5 to form enough p-n heterojunction, and therefore improve the photocatalytic H2 evolution activity.

3.6.5. Comparison of the Best Activities of the Four Types of Photocatalysts

The best activities of the four types of photocatalysts are listed in Table 2. The photocatalytic H2 evolution activities of TiO2-Y2Cu2O5 prepared by a sol-gel method and solid state method, respectively, Pt-Y2Cu2O5 and Y2Cu2O5 are in the order of TiO2-Y2Cu2O5 (sol-gel method) > Pt-Y2Cu2O5 > TiO2-Y2Cu2O5 (solid state method) > Y2Cu2O5. It can be concluded that Pt and TiO2 significantly enhanced photocatalytic activity of Y2Cu2O5. Pt-Y2Cu2O5 photocatalyst exhibits higher photocatalytic activity compared with TiO2-Y2Cu2O5 (solid state method), but lower than that of TiO2-Y2Cu2O5 (sol-gel method), indicating that composite method has a certain effect on photocatalytic activity of TiO2-Y2Cu2O5.

3.7. Durability of the Y2Cu2O5 Photocatalyst

The results of the durability for the Y2Cu2O5 photocatalyst are depicted in Figure 9. After four cycles of photocatalytic H2 evolution, the photocatalyst exhibits a little loss of activity, indicating that it is relatively unstable during the light photocatalysis process. In the first run of 12 h irradiation, the photocatalytic activity of Y2Cu2O5 achieves the maximum value at 6 h and then decreases little because of the continuous consumption of the sacrificial reagent oxalic acid. However, with the supplement of new oxalic acid, a little increase of the photocatalytic H2 evolution activity appears in the third run. But the amount of H2 generated in the 4th run is relatively lower than the others because a small Y2Cu2O5 will become Y2(C2O4)3·2H2O during the long photocatalytic reaction process.

4. Conclusions

In summary, Y2Cu2O5 has been successfully prepared via solid state reaction. The as-prepared photocatalyst shows good photocatalytic H2 evolution activity from aqueous oxalic acid solution when the photocatalyst concentration is 0.8 g L−1 under simulated sunlight irradiation. Pt-loaded Y2Cu2O5 shows higher H2 evolution activity as compared to the pure Y2Cu2O5. Furthermore, the combination of TiO2 and Y2Cu2O5 can effectively enhance the photocatalytic activity, and the photocatalytic H2 evolution activity (4.35 mmol h−1 g−1) is obtained for TiO2-Y2Cu2O5 with 30 wt.% TiO2 content prepared through a sol-gel approach. The as-prepared Y2Cu2O5 is a promising photocatalyst for the H2 evolution under sunlight irradiation.


This work was financially supported by National Nature Science Foundation of China (no. 21271071), the Natural Science Research Project of Hunan Province, China (no. 2012SK3174), the Natural Science Research Project of Yueyang City (no. 2012-3), and Aid Program for Science and Technology Innovative Research Team in Higher Educational Institutions of Hunan Province.