As an important catalytic material, CeO2 has attracted growing interest in various catalytic applications (for example, three-way catalysis, water-gas shift reaction, and catalytic wet air oxidation), due to its rich oxygen vacancies and low redox potential between Ce3+ and Ce4+ [1, 2, 3, 4]. To improve the catalytic performance, many techniques have been applied to Ce-containing materials, including ion doping, noble metal loading, and multi-metal oxide compositing [4, 5, 6]. In particular, noble metal loaded oxides have aroused much interest for application as heterogeneous cataclysts [1, 6, 7].
Recently, the strong metal-support interaction (SMSI) has been verified at the interface between noble metal and an oxides support, which dramatically changes the property of catalysts and favors their catalytic activity. For instance, introducing Au on the surface of ZnO greatly improves the catalytic activity for the CO oxidation by a SMSI interaction, via electron transfer from Au to the ZnO support [8]. Similarly, a Ce3+ content increase and catalytic activity enhancement were observed via the SMSI interaction by loading the catalyst with Pt [9], Pd [7], or Au [10] clusters on the surface of CeO2. In general, smaller particle size, larger interfaces, and good thermal stability favor the catalytic activity in a noble metal loaded CeO2 (e.g., Pt/CeO2) catalyst for CO oxidation [11].
Pd species, in the form of metallic Pd and cationic Pd, are of particular interest in CeO2 catalysis owing to their excellent catalytic performance. Due to SMSI interactions [12], cationic Pd species are generated by electron transfer from the Pd particles to the CeO2 support, and this results in a significantly enhanced activity for Pd/CeO2 in the synthesis of methanol from CO [13]. PdO has been reported to provide sites for CO chemisorption on surface, while PdO species in the CeO2 matrix generate oxygen vacancies that activate O2 in the catalytic oxidation of CO [14]. The chemisorption and catalytic performance of these Pd-O sites were also discussed by Colussi et al. [15].
Ce salts have been reported to have Fenton-like reactivity in the degradation of organic materials in aqueous solutions [16]. As an ambient catalytic reaction, a heterogeneous Fenton-like system, based on the CeO2 and H2O2, has been developed in our previous studies [17, 18, 19, 20, 21, 22, 23]. Similar to those occurring at high temperatures (> 600 °C) with CeO2 catalysis, as the Ce3+ content increases, the catalytic performance also improves for heterogeneous Fenton-like degradation reactions [18, 23]. For instance, Au nanocluster loading increases the Ce3+ content in CeO2 catalysts, thus promoting the Fenton-like catalytic activity of CeO2 in the degradation of acid orange 7 (AO7) [23].
PdO loading has improved the catalytic performance of high temperature (> 600 °C) CeO2 catalysts [14, 15]. We choose to investigate PdO/CeO2 for use in ambient temperature heterogeneous Fenton-like reactions. Therefore, PdO/CeO2 catalysts were prepared by deposition-precipitation method. The heterogeneous Fenton-like degradations of AO7 and salicylic acid (SA) were tested to reveal the role of PdO in the catalytic activity of CeO2 both in the dark and under visible irradiation.
Ce(NO3)3×6H2O and PdCl2 were purchased from Shanghai Sinopharm Chemical Reagent Co., Ltd in Shanghai, China. NH3×H2O (28%) and H2O2 (30%) were supplied by Shanghai Lingfeng Chemical Reagent Co., Ltd in Shanghai, China. AO7 dye and SA were obtained from Acros and Shanghai Lingfeng Chemical Reagent Co., Ltd, respectively. All reagents were of analytical grade and used as received without further purification. Ultra-pure water (18 MΩ/cm) was used throughout the experiments.
The precipitation method reported in our previous work [19] was employed to synthesize CeO2. NH3×H2O (6 mL, 28%) was added dropwise to a Ce(NO3)3 solution (0.1 mol/L, 150 mL) with vigorous stirring to form a yellowish suspension. The suspension was stirred continuously for 2 h at 40 °C followed by aging for 12 h at 40 °C. The resulting purple precipitate was collected by centrifugation, washed several times with water, and dried at 60 °C under vacuum for 12 h. After grinding, the precipitate was heated to 500 °C at 2 °C/min before being calcined at this temperature in air for 4 h to obtain the raw CeO2 powder.
The PdO/CeO2 catalyst was prepared by deposition- precipitation (DP) [24]. Raw CeO2 powder (0.86 g) was dispersed in 100 mL H2O, and the desired amount of PdCl2 was added to the suspension with vigorous stirring. Palladium hydroxide was precipitated on the CeO2 by the addition of 2.0 mol/L NaOH, maintaining the pH of the solution at 8.5. During the precipitation process, the suspension was stirred for 2 h at 70 °C. The precipitation was washed several times with water and dried at 60 °C under vacuum for 12 h to obtain the PdO/CeO2. According to the amount of PdCl2, described by the molar percent of Pd used, the as-prepared catalysts were labeled as CeO2, 0.05 PdO/CeO2, 0.1 PdO/CeO2, 0.5 PdO/CeO2, 1.0 PdO/CeO2 and 2.0 PdO/CeO2, respectively.
X-ray diffraction (XRD) analysis of the catalysts was carried out with a Rigaku Ultima IV apparatus using Cu Kα radiation (λ = 0.15406 nm) and a graphite monochromator at room temperature, operated at 40 kV and 40 mA. Diffraction patterns were recorded in the 2θ range of 5°-80°. The BET specific surface areas of the samples were determined by N2 adsorption- desorption at -196 °C (Micromeritics ASAP 2020M). Samples were degassed at 180 °C for 10 h prior to the measurement. Raman measurements were performed at room temperature using a Renishaw Via+ Reflex Raman spectrometer with the excitation wavelength of 514.5 nm. The X-ray photoelectron spectroscopy (XPS) instrument was a Thermo Fisher ESCALAB 250Xi system with Al Kα radiation (photon energy 1361 eV), and calibrated internally by the carbon deposit C (1s) binding energy (BE) at 284.6 eV. The shift of the binding energy due to relative surface charging was corrected using the C 1s level at 284.6 eV as an internal standard. The morphology of the samples was observed by high-resolution transmission electron microscopy (HRTEM, JEOL JEM 2000-EX) with an electron beam operating at an accelerating voltage of 200 kV.
The catalytic activity of the catalysts was evaluated in the dark and under visible irradiation. To avoid any interference from organics desorption, all reactions were carried out in a pre-mixed mode [19], i.e., the CeO2 was first mixed with H2O2 before the addition of the organic substances. In the Fenton-like reaction, 0.025 g catalyst powder was mixed with 1.4 mL H2O2 (5%, prepared in advance) and the mixture was stirred for 5 min. Then, 50 mL of the aqueous organic solution (AO7 of 35 mg/L or SA of 20 mg/L, pH 4.0) was added into the mixture at the start of the reaction time. At the given time intervals, samples (3 mL) were taken from the mixture and immediately centrifuged. The supernatant solution was withdrawn and analyzed by recording the absorbance at 484 nm (for AO7) or 296 nm (for SA) with a UV-Vis spectrophotometer (Shimadzu UV-2450).
As for a typical visible light irradiated experiment (photo-Fenton reaction), a 500-W tungsten halogen lamp equipped with a UV cut-off filter (λ > 420 nm) was used as a visible light source. The lamp was cooled with flowing water in a quartz cylindrical jacket around the lamp, and ambient temperature was maintained during the whole reaction. The visible irradiation was introduced into the reactor at the start of the reaction time.
The XRD patterns of the as-prepared CeO2 and PdO/CeO2 powders are shown in Fig. 1. All the samples displayed the cubic CeO2 (fluorite structure, JCPDS 34-0349). No Pd-related signals were seen in the spectra. The average crystal sizes of the samples are ca. 17-18 nm according to the Scherrer equation, based on the (111) reflection of CeO2 [25]. Table 1 lists the average crystal sizes, specific surface areas, and pore volumes of the CeO2 and PdO-CeO2 powders. All samples have similar specific surface areas and pore volumes. This shows that PdO deposition has caused only small differences in the physical properties of the CeO2. Consequently, the crystal size and specific surface area cannot be a basis for any possible variation of their catalytic performances.
The morphologies of CeO2 and Pd/CeO2 were observed with HRTEM. As shown in Fig. 2, CeO2 has a grain size of ca. 15 nm, which is consistent with the XRD results. The interplanar spacing (d) was 0.27 nm, which corresponds to the (200) planes of the cubic cell of CeO2 [14]. In the HRTEM images of 1.0 PdO/CeO2 and 2.0 PdO/CeO2, some smaller grains of ca. 5 nm appear and have an interplanar spacing of 0.28 nm, which fits the (200) planes of the cubic cell of PdO (JCPDS 46-1211).
Generally, the chemical states, especially the surface chemical states, of CeO2 play an important role on the Ce3+/Ce4+ redox cycle and significantly affect the catalytic activity. Therefore, the chemical states of the PdO/CeO2 were investigated by Raman and XPS spectra.
The Raman spectra of the as-prepared catalysts (Fig. 3) were examined to investigate the oxygen vacancies in the CeO2 [26, 27, 28, 29]. The strong peak at 463 cm-1 is due to an F2g vibrational in a cubic fluorite lattice [10, 29, 30], while the bands around 593 cm-1 and below 300 cm-1 have been attributed to defect-induced modes [20, 23, 26, 28]. It is known that the oxygen vacancy in CeO2 is closely related to the presence of Ce3+ and thus is vital to the CeO2/H2O2 catalytic system [18, 20, 23]. Practically, the concentration of oxygen vacancy in CeO2 can be estimated by the ratio of the Raman peak intensity at 593 cm-1 to that at 463 cm-1, that is I593/I463. The I593/I463 values of the CeO2 and PdO/CeO2 powders are presented in Fig. 3(b). The PdO loading gradually increases I593/I463 from CeO2 to 2.0 PdO/CeO2. This increase in the I593/I463 ratio implies increased amounts of PdO enhances the oxygen vacancy levels in CeO2.
In addition, a slight peak appears at 834 cm-1for PdO/CeO2, which has been assigned to the stretching vibration of a peroxide-like (O22-) species [10, 29]. The increased intensity at 834 cm-1suggests that the formation of surface peroxide-like species is also promoted by increased PdO loading, which in turn increases the amount of oxygen vacancy in CeO2, and finally contributes to increase the Fenton-like reactivity of these PdO/CeO2 composites.
The surface chemical compositions of the CeO2 and PdO-CeO2 (Fig. 4) were further investigated with XPS spectroscopy. Figure 4(a) presents the fine XPS spectra of Pd 3d. Although the Pd 3d signals can hardly be observed at the lower Pd contents, they are suitably intense at higher contents. As shown in Fig. 4(a), two characteristic peaks were observed for the PdO/CeO2 powders with a Pd content higher than 0.5 at%. According to the Ref. [7], the peaks at 343 and 338 eV correspond to the 3d3/2 and 3d5/2 signals of Pd2+ species, respectively. The 3d3/2 and 3d5/2 signals for metallic Pd0, which appear at 335.3 and 340.2 eV, respectively, were not observed in any of the PdO/CeO2 samples. The chemical stability of Pd species was also observed by re-measuring the XPS after the Fenton-like reaction for the degradation of AO7, where the same Pd2+ signals were observed.
Figure 4(b) shows the Ce 3d fine XPS spectrum of CeO2 and its corresponding deconvolution peaks. The deconvolution of Ce 3d spectrum has been labeled as two pairs of doublets (v0 /u0 and v’/u’) assigned to Ce3+ and three pairs of doublets (v/u, v’’/u’’ and v’’’/u’’’) assigned to Ce4+, in which vn and un refer to the 3d5/2 and 3d3/2 spin-orbit components, respectively [30, 31, 32]. The relative concentration of Ce3+ in the catalysts could be obtained by calculating the relative areas under the curve of each deconvoluted peak according to Eq. (1) [18, 20, 23, 33, 34]:
in which Ai is the integration area of peak i.
Figure 4(c) presents the calculated relative concentrations of Ce3+ in the CeO2 and PdO/CeO2 powders. The concentration of Ce3+ in CeO2 is 24.6%, less than in the PdO-deposited CeO2, which were 25.5%, 26.4%, 27.8%, 28.4% and 26.4% for 0.05 PdO/CeO2, 0.1 PdO/CeO2, 0.5 PdO/CeO2, 1.0 PdO/CeO2, and 2.0 PdO/CeO2, respectively. The PdO loading appears to have a direct effect on the chemical state of cerium ion in CeO2. The relative concentration of Ce3+ increases with increasing PdO loading, with a slight decrease at a PdO loading of 2.0 at%.
The XPS and the Raman results reveal that the concentrations of Ce3+ and the oxygen vacancy gradually increase with increased amounts of PdO. The enhanced Ce3+ (and the associated oxygen vacancy) content in the catalyst should significantly promote the redox catalytic performance of PdO/CeO2 [2]. By analogy, its activity of catalyzing H2O2 in the oxidation of organics should also be promoted.
The degradation of AO7 and SA was employed to evaluate the Fenton-like reactivity of the CeO2/H2O2 system. Figure 5 shows the Fenton-like degradation of AO7 in the dark and under visible light irradiation with various catalysts. To avoid any interference due to AO7 desorption, all reactions were carried out in a pre-mixed mode [19], i.e., CeO2 was first mixed with H2O2 before the addition of the organic substances.
As shown in Fig. 5, the Fenton-like degradation of AO7 was well fitted by first order kinetics. The corresponding degradation rate constants of AO7 of the CeO2 and PdO-CeO2 catalysts are listed in Table 2. Figure 5(a) shows the Fenton-like degradation of AO7 in the dark. The PdO loading obviously promoted the Fenton-like reactivity of the catalysts. The catalytic activity of the PdO/CeO2 increases with increased PdO loading and reaches a maximum at a loading of 1.0 at%. The Fenton-like reactivity rate constant of 1.0 PdO/CeO2 is 7.8-fold that of CeO2 in this reaction. The catalytic activity correlates with the Ce3+ concentration in the catalysts. Previous work suggests that the catalytic action of the CeO2/H2O2 system is mostly centered on the surface Ce3+ sites [35]. This Fenton-like degradation of AO7 gives additional support to thesurface Ce3+ site centered proposal for the CeO2/H2O2 system.
Figure 5(b) shows the photo-Fenton-like degradation of AO7 under visible irradiation. Under irradiation, the adsorbed AO7 can be excited by visible photons and inject an electron into the CeO2. Our previous work [17, 18, 23] suggested that the injected electron would be trapped by interfacial peroxide-like species on the CeO2, which would initiate an intramolecular rearrangement, and result in the release of free Ce3+ along with the produce of an ×OH radical (Eq. (2)). As free Ce3+ and an ×OH radical improve the Fenton-like reactivity, the degradation of AO7 is greatly accelerated by visible irradiation.
The rate constants of the visible light assisted photo- Fenton-like degradation of AO7 with the CeO2 and PdO-CeO2 catalysts are obviously higher than that without visible light irradiation. 1.0 PdO/CeO2 also exhibits the highest visible light assisted photo Fenton-like reactivity among the catalysts described here. The introduction of visible irradiation increases the AO7 degradation rate constants by 10.2 and 6.4 times with CeO2 and 1.0 PdO/CeO2, respectively. Notably, the combined effects of PdO loading and visible light irradiation accelerate the Fenton-like degradation of AO7 to ca. 50 times that of CeO2. It has been reported that the strong interaction between PdO and CeO2 would induce the generation of oxygen vacancies and Ce3+ as well in the CeO2, while dye sensitization accelerates the Ce3+ regeneration through interfacial peroxides species by interfacial electron injection [14]. As a Ce3+ site-centered interfacial reaction, the degradation of AO7 is thus greatly promoted.
A colorless chemical, SA, was also employed to evaluate the Fenton-like catalytic activity of PdO/CeO2. Figure 6 shows the Fenton-like degradation of SA with CeO2 and PdO-CeO2 catalysts in the dark and under visible irradiation. The corresponding Fenton-like degradation rate constants of SA are listed in Table 3. The PdO loading also affects the degradation rate of SA, and PdO/CeO2 with a PdO loading of 1.0 at% exhibits the highest catalytic activity in either case. However, the magnitude of the improvement is smaller than that for the Fenton-like degradation of AO7. Figure 6(b) shows the photo Fenton-like degradation of SA under visible irradiation. It seems that the introduction of visible irradiation has no effect on the Fenton-like degradation of SA. Since SA is a colorless substance, it is incapable of absorbing the visible photons and therefore does not help the Fenton-like degradation of SA in the CeO2/H2O2 system.
Unexpectedly, the catalytic activity of PdO/CeO2 for the degradation of SA decreases at low PdO loadings as the PdO loading increases, which was not seen for the degradation of AO7. At higher PdO loading, the catalytic activity of PdO/CeO2 increases to a maximum at 1.0 at%. The catalytic improvement by PdO loading is mainly attributed to the higher concentration of surface Ce3+ and additional interfacial peroxide species. However, according to our previous work [18, 19, 20, 23], the CeO2-based Fenton-like degradation of organics occurs via an interfacial reaction, and is an adsorption-triggered process; therefore, the adsorbability of the CeO2 should also affect the Fenton-like reactivity.
The adsorption rates of AO7 and SA are presented in Fig. 7. To avoid the possible interference from competitive adsorption by H2O2, the adsorption rates were measured according to the following procedure. An AO7 (or SA) solution was mixed with the corresponding catalyst in the dark for 1.0 h to achieve a pre-adsorption. Then, the required amount of H2O2 solution was quickly added to reach a concentration of 20 mmol/L. After stirring for 2 min and centrifuging to remove the catalyst, the concentration of AO7 (or SA) was measured to estimate the adsorption rate of the organic material.
Figure 7 shows that AO7 adsorbs well on the CeO2 and gives an adsorption rate of ca. 50%. The PdO loading does not significantly affect the absorbability of the catalysts. However, the adsorption of SA on CeO2 is significantly weaker than for AO7, which gives an adsorption rate of 13% on CeO2. The PdO loading further reduces this at lower concentrations. As shown in Fig. 7, there are only 7% and 3% of SA molecules adsorbed on the 0.05 PdO/CeO2 and 0.1 PdO/CeO2, respectively. Beyond this, the adsorption increases to close to the initial value. As the Fenton-like degradation of organic material based on CeO2/H2O2 is an interfacial reaction, the degradation kinetics of the organic material depends on its adsorption onto the surface of CeO2. At low PdO contents (less than the 0.5 at%), the deposited PdO on the surface of CeO2 occupies the SA adsorption sites, thus decreasing the ability of the catalysts to adsorb SA and this decreases the catalytic activity. As the PdO content increases further, the adsorption of SA on the catalysts is restored. Simultaneously, as the interaction between PdO and CeO2 increases, the Ce3+ content in the CeO2 increases. The Fenton-like degradation of SA is significantly accelerated, reaching a maximum at the PdO loading of 1.0 at%.
PdO/CeO2 catalysts were prepared by deposition- precipitation method. The PdO/CeO2 composites have a cubic CeO2 structure, ca. 17 nm in diameter and have similar specific surface areas with that of bare CeO2. The Pd in the composites is present as Pd2+, and the deposited PdO particles are ca. 5 nm in size. The introduction of PdO increases the relative concentration of Ce3+ in CeO2. Deposition of PdO accelerates the Fenton-like degradation of SA, with a maximum rate at the PdO loading of 1.0 at%. For AO7, a dye sensitization effect was observed under the visible irradiation, which accelerates the regeneration of Ce3+ from interfacial peroxides species by interfacial electron injection from an excited AO7 molecule to the CeO2 catalyst. The Fenton-like degradation rate constant of AO7 with bare CeO2 in dark is 0.078 h-1, while that with 1.0 PdO/CeO2 increases to 3.9 h-1 under visible irradiation, a 50-fold improvement.
2015, Vol. 36 Issue (3): 314-321
PDF (1024 KB)
2015, Vol. 36 Issue (3): 314-321
PDF (1024 KB)