Synthesis of TiO2/Pd and TiO2/PdO Hollow Spheres and Their Visible Light Photocatalytic Activity

A series of TiO2, TiO2/Pd, and TiO2/PdO hollow sphere photocatalysts was successfully prepared via a combination of hydrothermal, sol-immobilization, and calcination methods. The structure and optical properties of the as-prepared samples were characterized by X-ray diffraction, field emission scanning electron microscopy, transmission electron microscopy, Brunauer-Emmett-Telleranalysis, Barrett-Joyner-Halenda measurement, and UV-Vis diffuse reflectance spectroscopy. The photocatalysis efficiencies of all samples were evaluated through the photocatalytic degradation of rhodamine B under visible light irradiation. Results indicated that TiO2/PdO demonstrated a higher photocatalytic activity (the photocatalytic degradation efficiency could reach up to 100% within 40min) than the other samples and could maintain a stable photocatalytic degradation efficiency for at least four cycles. Finally, after using different scavengers, superoxide and hydroxyl radicals were identified as the primary active species for the effectiveness of the TiO2/PdO photocatalyst.


Introduction
Semiconductor materials have been received and extensively applied as photocatalysts in photocatalytic water splitting and photodegradation of organic pollutants. TiO 2 is an important semiconductor in wide applications, such as in photocatalysts [1,2], solar cells [3], electrorheology [4,5], and antibacterial agents [6]. It has been exhaustively explored for its merits of environment friendliness, chemical inertness, high stability, nontoxicity, and low cost when utilized as a photocatalyst for environment remediation and photodegradation of organic pollutants [7][8][9].
Xiang et al. [10] successfully synthesized 3D urchin-like TiO 2 microspheres for photocatalytic applications. Lai et al. [11] prepared hierarchical TiO 2 and hierarchical nanosheetassembled yolk-shell TiO 2 microspheres for dye adsorption and photocatalysis, respectively. Sarkar and Chattopadhyay [12] synthesized a large-scale mesoporous TiO 2 microsphere for photocatalysis degradation. However, these TiO 2 photocatalysts exhibit high photocatalytic activity only under UV irradiation. This finding can be ascribed to the intrinsic prop-erties of TiO 2 : a wide band gap (3.2 eV), which limits its use in the UV region of the solar spectrum, which accounts for a very small fraction of solar energy [13]. Therefore significant efforts have been devoted to modify the band gap of TiO 2 . Doping to extend the optical absorption of TiO 2 into the visible spectrum is an effective method. In recent decades, doping with metals or nonmetals has been performed to shift the optical response of the catalytically active TiO 2 from the UV to the visible light region. For example, mordenite [14], graphene [15,16], F [17], Sm [18], Ni [19], Pt [20], Au [21], Ag [22], and Pd [23,24] have been used to extend the photoresponse range of TiO 2 . Similar to Pd, doped PdO can form a certain number of heterojunctions with TiO 2 , thereby inhibiting the recombination of photogenerated carriers, improving the light utilization efficiency and catalytic activity.
In this study, we successfully synthesized TiO 2 /Pd and TiO 2 /PdO hollow sphere photocatalysts following the methods of Li et al. [25]. The methods combined hydrothermal, sol-immobilization, and calcination. We selected rhodamine B (Rh B) as the dye template for studying the photocatalytic degradation efficiency of the as-prepared samples under visible light [26]. The results exhibited that TiO 2 /PdO demonstrated a higher photocatalytic activity than previously reported photocatalysts [27,28] because TiO 2 /PdO possessed a wide range of optical absorption, high crystallinity, and a large pore size. Moreover, the photocatalytic reaction mechanism was also discussed on the basis of trapping experiments of active species.
2.2. Synthesis of TiO 2 Hollow Spheres. We followed the methods of Li et al. to prepare the TiO 2 hollow spheres [25]. In a typical procedure, a homogeneous solution including 51 mL of H 2 O, 14.1 mL of DMF, and 6.9 mL of TFA was dropped to 2.04 mL of TBOT in a Teflon-lined autoclave, and then the mixture solution was heated to 180°C and maintained this temperature for 24 h. After cooling to room temperature, the white precipitate was obtained and washed with absolute ethanol and distilled water three times, respectively. Then, the obtained TiO 2 hollow spheres were dried in a vacuum at 60°C and calcined at 450°C for 3 h in air. The TiO 2 hollow spheres before and after calcination were denoted TiO 2 and TiO 2 -T, respectively.
2.3. Synthesis of TiO 2 /Pd and TiO 2 /PdO Hollow Spheres. Pd nanoparticles were supported on TiO 2 hollow spheres by using the methods of Li et al. [25]. An aqueous solution of PdCl 2 (10 mL) and polyvinyl alcohol (PVA, 1 wt%, 0.72 mL) were diluted with water (120 mL). After stirring for 30 min, 1.98 mL NaBH 4 solution (0.1 M) was added to form a dark brown solution. Subsequently, sulfuric acid solution was added to adjust the pH to 1. Then, 0.2 g TiO 2 was ultrasonically dispersed in this mixture and rapidly stirred for 2 hours. The resulting slurry was filtered, washed with absolute alcohol, and distilled water, respectively. The obtained TiO 2 /Pd hollow spheres were dried in a vacuum at 60°C and calcined at 450°C for 3 h in air. The TiO 2 /Pd hollow spheres before and after calcination were denoted TiO 2 /Pd and TiO 2 /PdO, respectively.

2.4.
Characterization. X-ray diffraction (XRD, Philips X'Pert PRO, Netherlands) on a diffractometer equipped with Cu Kα radiation (k = 0:15418 nm) from 20°to 80°at a scanning speed of 2°min -1 . The morphology and detailed structural information of the samples were observed using a field emission scanning electron microscope (FESEM, Ultra 55, CARL ZEISS in Germany) and transmission electron microscopy (TEM, Libra 200 PE, operated at 200 kV, Germany). The nitrogen adsorption-desorption isotherms were measured at 77 K using a Micromeritics ASAP 2000 system, and the surface area and pore size distribution were calculated using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods, respectively. The optical spectra were examined by using a Hitachi U-4100 UV-Vis diffuse reflectance spectra (UV-Vis DRS) in the wavelength range of 200-800 nm equipped with the integrating sphere accessory for diffuse reflectance spectra.
2.5. Photocatalytic Activity Measurement. Photocatalytic activities of the as-synthesized products were studied using Rh B as a model pollutant under visible light irradiation. The photocatalytic reactor (Beijing NBT Technology Co., Ltd.) consisted of quartz glass with a circulating water jack and a light source (A 500 W xenon lamp). There is circulating water between the reaction system and the xenon lamp to remove the thermal effect of light. The experiments were performed at room temperature. For each photocatalytic activity measurement, 30 mg of photocatalyst was added to 50 mL of Rh B aqueous solution with the concentration of 30 mg/L. Before irradiation, the reaction mixture was stirred in the dark at 25°C for 20 min to ensure the establishment of absorption/desorption equilibrium between the photocatalysts and Rh B solution. After the absorption equilibrium, the suspension solution was irradiated by the visible light with continuous stirring. At the given intervals, 5 mL of reaction mixture was centrifuged and then analyzed by recording variations of the maximum absorption spectra by UV1600 spectrometer [29].         Figure 4 presents the nitrogen adsorption-desorption isotherms and pore size distribution curve of TiO 2 , TiO 2 -T, TiO 2 /Pd, and TiO 2 /PdO. The isotherms of these samples could be attributed to type IV, and the curves exhibited hysteresis loops at high relative pressures, indicating a mesoporous structure. Furthermore, the specific surface areas and average pore sizes of these samples are summarized in Table 1 . This result may be because the average pore size is related to the crystallite size of a sample [33]. When the hollow spheres were calcining, the crystallites located in the inner cores were easier to dissolve than the exterior 5 International Journal of Photoenergy nanoparticles, the exterior nanoparticles could serve as the new starting site for the subsequent recrystallization process, and continued crystal growth could lead to the increased crystallite size of the sample [30]. Therefore, the average pore sizes of TiO 2 -T and TiO 2 /PdO after calcinations were larger than those of TiO 2 and TiO 2 /Pd without calcination. The specific surface area of the catalyst is mainly determined by the pore wall area in the crystal. Due to the hightemperature calcination process, the hollow spheres dissolved and grew, which would cause the pores in the crystal to collapse and lose fine voids. Therefore, the BET specific surface areas of the calcined TiO 2 -T and TiO 2 /PdO were smaller than that of the uncalcined TiO 2 and TiO 2 /PdO [34].

UV-Vis Diffusive Reflectance Spectra (UV-Vis DRS)
Analysis. Figure 5 shows the UV-Vis adsorption spectra of TiO 2 , TiO 2 -T, TiO 2 /Pd, and TiO 2 /PdO. The corresponding absorption edge wavelengths and band gaps are summarized in Table 2. The band gap energy can be estimated by using the equation E g = 1240/λ g , where E g is the band gap energy (eV) and λ g is the wavelength (nm) of the absorption edge in the spectrum, respectively. The absorption spectra of TiO 2 and TiO 2 -T were only in the ultraviolet region and can be ascribed to the intrinsic band gap absorption of TiO 2 . Compared with those of TiO 2 and TiO 2 -T, the absorption spectra of TiO 2 /Pd and TiO 2 /PdO showed more absorption in the visible range (400-700 nm). The absorption intensities in the visible light region decreased in the order of TiO 2 /PdO>TiO 2 /Pd>TiO 2 -T>TiO 2 . In other words, the band gap energies decreased in the order of TiO 2 >TiO 2 -T>TiO 2 /Pd>TiO 2 /PdO. Numerous previous studies indicated that TiO 2 doping with a noble metal can effectively increase absorption and significantly enhance the visible light photocatalytic properties of TiO 2 [35].

Photocatalytic Activity and Recycled Photocatalytic
Degradation Performance. The photocatalytic activities of TiO 2 , TiO 2 -T, TiO 2 /Pd, and TiO 2 /PdO were evaluated in terms of the decolorization of Rh B under visible light irradiation as shown in Figure 6(a). The photocatalytic performances of a blank Rh B solution without photocatalysts and TiO 2 /PdO without visible light irradiation are also included in Figure 6(a) for comparison. The chart clearly shows that the photocatalytic degradation rate of Rh B increased with the intensity of irradiation and the relative photocatalytic activity of the catalysts decreased in the order TiO 2 /PdO>TiO 2 /Pd>TiO 2 -T>TiO 2 >Bare>TiO 2 /PdO (black) ( Table 3). First, the final photocatalytic degradation rate of Rh B during self-degradation and adsorption was lower than that under visible light and with the presence of photocatalysts, thereby revealing the photocatalytic activity of these samples. Second, TiO 2 /PdO and TiO 2 /Pd exhibited higher activities than TiO 2 and TiO 2 -T. The higher activities of TiO 2 /PdO and TiO 2 /Pd could be explained in terms of the enhancement of UV-Vis absorbance spectra due to Pd and PdO doping (as shown in Figure 5) [35]. The enhancement of UV-Vis absorbance spectra of the doped samples represents their optical absorption property, offering a higher photocatalytic activity. Third, the visible light photocatalytic activities of TiO 2 /PdO and TiO 2 -T after calcination were higher than those before calcination (TiO 2 and TiO 2 /Pd). This finding could be ascribed to the crystallinity of the samples; that is, high crystallinity indicates few defects, which could lead to the slowed recombination of photogenerated hole-electron pairs [36]. The crystallinity of TiO 2 -T and TiO 2 /PdO was improved by calcination (as shown in Figure 1). Thus, the photocatalytic activities of TiO 2 /PdO and TiO 2 -T were better than those of TiO 2 and TiO 2 /Pd.     International Journal of Photoenergy The UV-Vis spectra changes in the dye solution over various time intervals are shown in Figure 6(b) to understand and clarify the changes in the molecular structure of Rh B in the presence of TiO 2 /PdO under visible light irradiation. The absorption peak of 550 nm continuously descended with increasing exposure time and almost vanished after 40 min of irradiation. This result is consistent with the gradual color change of the reaction solution from peach to almost colorless during the different reaction times, which suggests that TiO 2 /PdO exhibited an excellent photocatalytic activity for Rh B degradation. In addition, most dye degradation processes occurred in the initial minutes. Immediately after 10 min, TiO 2 /PdO obtained a degradation percentage of 81.7%. Therefore, the degradation rate for TiO 2 /PdO in the initial minutes was high.
The kinetics of the photodegradation of Rh B in the presence of TiO 2 , TiO 2 -T, TiO 2 /Pd, and TiO 2 /PdO was also studied. The −ln ðC/C 0 Þ of these samples presented a good linear reaction with irradiation time (shown in Figure 6(c)), which means that the photodegradation of Rh B follows the rules of the first-order reaction kinetics −ln ðC/C 0 Þ = kt, where k is the first-order rate constant, and C 0 and C are the concentrations of Rh B at irradiation times 0 and t, respectively [11]. The calculated kinetic rate constants are summarized in Table 3. TiO 2 /PdO had a maximum k value of 0.126 min −1 , whereas the others had 0.034 min -1 (TiO 2 ), 0.058 min -1 (TiO 2 -T), 0.063 min -1 (TiO 2 /Pd), 0.0079 min -1 (Bare), and 0.00026 min -1 (TiO 2 /PdO black). Moreover, the rate constant coincides with the final degradation rate (Table 3) of the photocatalytic degradation in Figure 6(a).
The degradation (%) and rate constant ðkÞ of previously reported photocatalysts, such as Pd/TiO 2 [27], PdO/TiO 2 [27], TiO 2 [12], and P25 [11], are listed in Table 3 for comparison. Notably, the degradation and rate constant of TiO 2 /PdO were determined to be 100% and 0.126 min -1 , which were much larger than those of PdO/TiO 2 and other photocatalysts. The high degradation and rate constant of TiO 2 /PdO could be attributed to its wide range of optical absorption, high crystallinity, and large pore size.
Recycling of catalysts is a key step in assessing the practical application of photocatalysts. To confirm the possibility of recycling the obtained photocatalysts, four recycling experiments were performed with a TiO 2 /PdO sample in the photocatalytic degradation of Rh B. As shown in Figure 6(d), the degradation rates of four recycling instances reached up to 99.8%, 99.0%, 98.4%, and 96.6%. The results reveal that the separation of photocatalysts was effective. Thus, the photocatalyst is basically stable and therefore promising for environmental remediation.

Active Species and Mechanism
Underlying the Photocatalytic Degradation of Rh B. The effect of different scavengers on the degradation of Rh B by the TiO 2 /PdO was investigated to confirm the main active species and provide insight into the potential photocatalytic mechanism of TiO 2 /PdO photocatalysts. Three different scavengers, namely, BQ (a ·O 2 radical scavenger), IPA (a ·OH radical scavenger), and KI (h + scavenger), were introduced into the reaction systems [37,38]. Figure 7 shows that BQ markedly suppressed the degradation rate of Rh B, whereas the presence of KI barely inhibits the photocatalytic efficiency. Compared with KI, IPA hindered the degradation of Rh B. Thus, we conducted a set of comparative experiments. Both BQ and IPA scavengers were added to the same set of experiments to obtain the results shown in Figure 7. IPA+BQ inhibited the degradation of Rh B more strongly than BQ. Therefore, these results suggest that ·OH and ·O 2 were the primary active species that governed the photocatalytic process, and h + was not the main active species in the degradation of Rh B by the TiO 2 /PdO.
On the basis of the above investigation and analysis, we propose the possible mechanism for the charge transfer and separation processes underlying the photocatalytic degradation of TiO 2 /PdO as schematically illustrated in Figure 8. Upon visible light irradiation, PdO absorbs incident photons and generates electron-hole pairs:  7 International Journal of Photoenergy where e − CB and h + VB are the electrons in the conduction band (CB) and the holes in the valence band (VB), respectively. The photogenerated electrons transfer from the CB of PdO to the CB of TiO 2 , while the photogenerated holes remain in the VB of PdO. Thus, the electrons and holes can be effectively separated to inhibit their recombination [39]. Finally, the electrons reduce O 2 to ·O 2 radicals; the holes can react easily with the surface-bound H 2 O and generate ·OH radicals. The powerful oxidizing agent (·O 2 and ·OH) can then react with the organic dyes and produce CO 2 , H 2 O, and so on [12].

Conclusion
We have successfully synthesized TiO 2 /Pd and TiO 2 /PdO hollow spheres. The photocatalysis efficiencies of all asprepared samples were evaluated through the photocatalytic degradation of Rh B under visible light irradiation. The results indicated that TiO 2 /PdO demonstrated a higher photocatalytic activity than the other samples, and a Rh B of 30 mg/L could be completely (100%) degraded by TiO 2 /PdO under visible light irradiation within 40 min. This result can be ascribed to the wide-range optical absorption, high crystallinity, and large pore size of TiO 2/ PdO. In addition, TiO 2 /PdO can be easily separated and retain its photocatalytic efficiency for at least four cycles of reaction under visible light irradiation. Thus, TiO 2 /PdO is a promising photocatalyst for the decomposition of recalcitrant and emerging pollutants in water and wastewater treatments.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare that they have no conflict of interest.