Volatile organic compounds (VOCs), sulfur oxides, nitrogen oxides, carbon monoxide and particulate matter are all harmful to the environment and to human health. Many countries have issued increasingly strict regulations restricting the emissions of VOCs in the 21st century. Among the different strategies used to control VOC emissions, catalytic combustion has proven to be an effective technology. One key step in the further advancement of this technology is the development of highly efficient catalysts.
In recent years, supported transition metal and noble metal catalysts have received much attention because these materials have several advantages, especially with regard to catalytic performance, compared with supported noble metal or transition metal catalysts. Among the supported noble metal catalysts, Pt, Pd or Rh on high surface area materials have been widely used due to the exceptional low-temperature activity and recyclability of such systems during the abatement of VOCs. However, these catalyst systems tend to exhibit low resistance to poisoning and are also costly compared with transition metal oxides [1, 2]. Because catalytic activity varies depending on both the noble metal and the particular VOC, different activities are expected over different supported noble metal catalysts when applied to specific reactions [3]. For example, supported Rh catalysts are the most active for hexene oxidation but perform poorly during the oxidation of aromatics [3]. Supported Pd is more active in the oxidation of short-chain hydrocarbons, whereas supported Pt exhibits higher activity for the oxidation of long-chain hydrocarbons and aromatics [4, 5]. Li et al. [6] prepared Pt/γ-Al2O3 catalysts using a self-assembly method and observed that the temperatures required for the complete oxidation of toluene, isopropanol, acetone and ethyl acetate over 1.0% Pt/γ-Al2O3 were 130,135,145 and 215 ℃, respectively, at a space velocity (SV) of 18000 mL g-1 h-1. They proposed that the high catalytic activity of 1.0% Pt/γ-Al2O3 is attributable to its enhanced surface area, and small size and good dispersion of Pt nanoparticles (NPs), reproducible low-temperature activity and greater concentration of hydroxyl species. Upon loading Pt NPs 1.3-2.3 nm in size on the surface of ZSM-5, Chen et al. [7] found that a Pt/ZSM-5 catalyst with a Pt particle size of 1.9 nm gave the best performance (T98% = 150 ℃ at SV = 30000 mL g-1 h-1) during the complete oxidation of toluene. This result was attributed to an optimum Pt dispersion and an ideal proportion of Pt0 species.
Supported transition metal catalysts not only show high oxidation reaction activities but are also less expensive and more resistant to poisoning compared with supported noble metal catalysts [8, 9]. To date, numerous transition metal oxides, such as CuO, Co3O4, Cr2O3, NiO, Fe2O3 and MnO2 [10], have been reported to be active for the catalytic combustion of typical VOCs. Among these, cobalt oxides exhibit excellent catalytic activity and selectivity [11]. It is also well known that the nature and extent of dispersion of the cobalt oxide species are the main factors determining the performance of these catalysts [12, 13]. The preparation method [14, 15], type of support [16-18] and transition metal loading [15, 19] can all affect the state of the cobalt oxide species. Furthermore, the addition of a metal oxide can enhance the catalytic performance of supported platinum catalysts. Liu et al. [20] prepared Y- and Mn-doped Pt/Ce0.50Zr0.50O2/γ-Al2O3 catalysts and observed that 0.5% Pt/50% Ce0.40Zr0.40Y0.10Mn0.10Ox/γ-Al2O3 exhibited the highest activity (T10% = 170 ℃ and T90% = 216 ℃ at SV = 12000 h−1) during toluene combustion. Using a plasma- and ultrasound-assisted wet impregnation strategy, Haghighi and coworkers [21] found that an as-prepared 1% Pt/30 wt% CeO2-Al2O3 catalyst showed good low-temperature activity (T100% = 180 ℃ at SV = 8400 mL g-1 h-1) and high stability during toluene oxidation. In these previous studies, porous alumina has been commonly used as the support material. To the best of our knowledge, however, there have been no reports concerning the preparation of three-dimensionally ordered macro-/mesoporous alumina (3DOM Al2O3)-supported Co3O4 and Pt NPs and their catalytic application to the combustion of toluene.
Recently, our group has prepared numerous 3DOM-supported transition metal oxide and/or noble metal catalysts, including MnOx/3DOM LaMnO3 [22], Co3O4/3DOM La0.6Sr0.4CoO3 [23], Au/MnOx/3DOM La0.6Sr0.4MnO3 [24] and Au/MnOx/3DOM SiO2 [25]. It has been found that these porous materials perform well with regard to catalyzing the oxidation of CO and/or typical VOCs. We herein report the preparation, characterization and catalytic properties of xPt/yCo3O4/3DOM Al2O3 (Pt mass fraction x% = 0-1.4%; Co3O4 mass fraction y% = 0-9.2%) during the combustion of toluene.
A well-packed, monodispersive colloidal crystal template composed of poly(methyl methacrylate) (PMMA) microspheres with an average diameter of approximately 300 nm was synthesized using a previously described approach [26]. The 3DOM Al2O3 with three-dimensionally ordered mesoporous walls was fabricated by employing the triblock copolymer Pluronic F127 (EO106PO70EO106) as a soft template and the PMMA microspheres as a hard template. It has been determined that the use of aluminum isopropoxide as the Al source tends to produce higher quality 3DOM Al2O3 [26], and so this reagent was applied during the preparation of the 3DOM Al2O3 in the present work. A typical preparation procedure began with the dissolution of 4.0 g F127 in 20.0 g of a 95% ethanol solution, and stirring until a homogeneous mixture was obtained (termed Solution A). Simultaneously,4.8 mL of concentrated nitric acid 65%-68% was added to 24.0 mL of absolute ethanol and the mixture was stirred for 15 min. Next,8.16 g of aluminum isopropoxide was gradually added to the above inorganic acid-ethanol solution until the Al precursor was completely dissolved, giving Solution B. Solution B was gradually poured into Solution A with continuous stirring and, after stirring for an additional 15 min, Solution C was obtained. Finally,4.0 g of the colloidal crystalline PMMA microspheres was added to Solution C. After the PMMA microspheres were thoroughly wetted, the excess solution was removed using a Buchner filter funnel connected to a vacuum. After drying under ambient conditions for 48 h, the obtained solid was calcined in an oven, applying a ramp of 1 ℃/min from room temperature (RT) to 300 ℃ and then holding the sample at this temperature for 3 h. Subsequently, the same ramp was applied to raise the temperature from 300 to 600 ℃, and the sample was held at 600 ℃ for 5 h, generating the 3DOM Al2O3 support.
The yCo3O4/3DOM Al2O3 sample with a theoretical value of y= 10 was prepared via incipient wetness impregnation. It should be noted that the 3DOM architecture would have been degraded when using water as the solvent to prepare the yCo3O4/3DOM Al2O3 during the subsequent thermal treatment process. To avoid this, we instead used isopropyl alcohol as the solvent during the fabrication of the yCo3O4/3DOM Al2O3. In a typical preparation procedure, the required amount of a 10% aqueous Co(NO3)2 solution was diluted with a specific volume of isopropyl alcohol, and the diluted solution was added dropwise to 0.2 g of the 3DOM Al2O3 support until the support was completely wetted. Following this, the wetted sample was held at RT for 0.5 h and then dried in an oven overnight at 80 ℃. In the final step, the resulting solid was calcined in air, applying a ramp of 1 ℃/min from RT to 500 ℃, and it was maintained at this temperature for 4 h, thus generating the yCo3O4/3DOM Al2O3.
The xPt/yCo3O4/3DOM Al2O3 specimens (with theoretical values of x = 0.5,1.0 and 2.0) were prepared using a polyvinyl alcohol (PVA)-protected reduction method [27]. In a typical preparation procedure, the required amount of PVA (Mr = 10000) was added to a 0.01 mol/L aqueous H2PtCl6 solution (Pt/PVA mass ratio = 1.0:1.5) in an ice-water bath under vigorous stirring over the span of 10 min. After rapidly injecting aqueous NaBH4 solution (0.1 mol/L, Pt/NaBH4 molar ratio = 1.0:5.0) and continuously stirring for 20 min, a sol was obtained. The necessary amount of the yCo3O4/3DOM Al2O3 support was subsequently added to a specific quantity of this sol, and the resulting suspension was sonicated at 60 kHz for 30 s. Nitrogen bubble-assisted stirring was then applied to agitate the suspension system for 10 h until complete adsorption of the colloidal noble metals. The solid product was then filtered, washed with 2.0 L of deionized water, dried at 80 ℃ for 12 h, and finally calcined at a ramp of 1 ℃/min from RT to 450 ℃ and maintained at this temperature for 4 h to produce the xPt/yCo3O4/3DOM Al2O3. Inductively coupled plasma atomic emission spectroscopy (ICP-AES, Thermo Electron IRIS Intrepid ER/S) analysis demonstrated that the actual Pt loading (x%) in the xPt/yCo3O4/3DOM Al2O3 ranged from 0.3% to 1.4%, while the actual Co3O4 loading levels (y%) were 9.2% in the Co3O4/3DOM Al2O3 and 8.9% in the xPt/yCo3O4/3DOM Al2O3.
To better compare the catalytic activities of the different materials, we also prepared a commercial Al2O3-supported Co3O4 and Pt sample. The commercial alumina (hydrophilic fumed alumina, Degussa AEROIDE Alu C, denoted as com-Al2O3) had a surface area of 110 m2/g. In addition, a 1.2Pt/8.8Co3O4/com-Al2O3 sample was prepared via the above incipient wetness impregnation and PVA-protected reduction routes. All the chemicals employed in this work were of A.R. grade purity and were purchased from the Beijing Sinopharm Chemical Reagents Company (Beijing, China) and used without further purification.
The actual Co3O4 and Pt concentrations in the samples were determined using ICP-AES. For these analyses, the samples were dissolved in a mixture of concentrated HCl and HNO3 (3:1 (V:V)). X-ray diffraction (XRD) patterns of the samples were recorded on a Bruker D8 Advance diffractometer with Cu Kα radiation and a nickel filter (λ = 0.15406 nm). Scanning electron microscopy (SEM) images of the samples were obtained with a Gemini Zeiss Supra 55 apparatus operating at 10 kV. Transmission electron microscopy (TEM) images and selected-area electron diffraction patterns of the samples were acquired using a JEOL-2010 instrument operating at 200 kV. The Brunauer-Emmett-Teller (BET) surface areas of samples were determined via N2 adsorption at −196 ℃, employing a Micromeritics Tristar 3000 analyzer, with degassing of specimens at 250 ℃ for 3 h under vacuum prior to measurements. X-ray photoelectron spectroscopy (XPS, VG CLAM 4 MCD analyzer) was used to determine the Co 2p, O 1s, Pt 4d, and C 1s binding energy (BE) of surface species, applying a Mg Kα (hv = 1253.6 eV) excitation source. Surface adsorbed water and carbonate species were removed by pretreating samples in O2 (flow rate = 20 mL/min) at 450 ℃ for 1 h followed by cooling to RT and transferring the pretreated samples into the spectrometer, all within a transparent glove bag (Instruments for Research and Industry, USA) filled with helium. The pretreated samples were outgassed in the preparation chamber of the instrument (at 1.33 mPa) for 0.5 h and then introduced into the analysis chamber (at 0.4 mPa) to acquire XPS spectra. The C 1s signal at 284.6 eV was used as a reference for BE calibration. Hydrogen temperature-programmed reduction (H2-TPR) experiments were carried out with a chemical adsorption analyzer (Autochem Ⅱ 2920, Micromeritics). Prior to TPR measurements, approximately 0.02 g of the catalyst (40−60 mesh) was loaded into a fixed-bed, U-shaped quartz microreactor (i.d. = 4 mm) and pretreated in an O2 flow of 30 mL/min at 300 ℃ for 1 h. After cooling to RT under the same atmosphere, the pretreated sample was exposed to a 50 mL/min flow of a 5% H2−95% Ar mixture and then heated from RT to 800 ℃ at a ramp of 10 ℃/min. Variations in the H2 concentration of the effluent were monitored via an on-line chemical adsorption analyzer. The reduction peak was calibrated by comparison with the peak obtained from the complete reduction of standard powdered CuO (99.995%, Aldrich).
The catalytic activities of the samples were evaluated in a continuous flow, fixed-bed quartz microreactor (i.d. = 4 mm). To minimize the effects of hot spots, each sample (50 mg,40-60 mesh) was diluted with 0.25 g of quartz sand (40-60 mesh). Prior to each test, the sample was treated in O2 (20 mL/min) at 250 ℃ for 1 h. After cooling to the desired temperature, the reactant gas (1000 ppm toluene + O2 + N2 (balance)) was passed through the catalyst bed at a total flow rate of 16.7 mL/min, equivalent to a toluene/O2 molar ratio of 1/400, and an SV of approximately 20000 mL g-1 h-1. The 1000 ppm toluene concentration in the gas mixture was generated by passing the N2 flow through a bottle containing pure toluene and chilled in an ice water bath. The toluene concentration in the gas stream was varied from 2000 to 3000 ppm by changing the N2 flow rate through the bottle. Water vapor was introduced into the stream at a level of 5.0% by passing the reactant mixture flow through a water saturator device held at 33 ℃. Both reactants and products exiting the reactor were analyzed online by a gas chromatograph (GC-2010, Shimadzu) equipped with a flame ionization detector and a thermal conductivity detector in conjunction with a Stabilwax@-DA column (30 m in length) for VOCs separation and a Carboxen 1000 column (3.175 mm in diameter and 3 m in length) for permanent gas separation. Products were also analyzed by mass spectrometry (Hiden HPR20), and it was determined that CO2 and H2O were the only products generated during the combustion of toluene. The carbon balance throughout the investigation was estimated to be (99 .5 ± 1.5)%.
Fig. 1 shows the XRD patterns of the as-prepared samples. By comparing these to the XRD pattern of the standard Al2O3 sample, it is evident that the three broad diffraction peaks at 2q = 26.0°,42.0° and 66.0° can be assigned to the amorphous Al2O3 phase in 1.4Pt/3DOM Al2O3. In addition, the peaks at 2q = 32.8°,34.5°,39.5°,45.6° and 67.0° are ascribable to the d-Al2O3 phase (JCPDS PDF #16-0394) in 1.2Pt/8.8Co3O4/com-Al2O3. Those at 2q = 19.0°,31.2°,36.8°,49.9°,55.7°,59.4°,65.1° and 77.2° result from the cubic Co3O4 phases (JCPDS PDF #42-1467) in 9.2Co3O4/3DOM Al2O3,0.3Pt/8.9Co3O4/3DOM Al2O3,0.6Pt/8.9Co3O4/3DOM Al2O3,1.3Pt/8.9Co3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3. Following loading of the Pt NPs, no peaks due to the Pt phase are seen in the xPt/yCo3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 patterns, indicating that the Pt NPs were highly dispersed on the Al2O3 support surfaces.
Fig. 2 presents SEM and TEM images of the 3DOM Al2O3 and 9.2Co3O4/3DOM Al2O3 samples. It can be clearly seen that the 3DOM Al2O3 displays a high-quality 3DOM architecture with a macropore diameter of 180-200 nm (Fig. 2(a) and (b)). Furthermore, the formation of ordered mesopores on the walls of the macropores is evident in Fig. 2(b). Additionally, the loading of Co3O4 NPs did not degrade the porous structure of the 3DOM Al2O3, and the Co3O4 NPs were uniformly dispersed (Fig. 2(c) and (d)). As shown in Fig. 3(a),(c),(e),(g) and (i), the high quality 3DOM structure contained interconnected macropores. It is worth pointing out that ordered mesopores with 4-6-nm diameters were generated within the macroporous skeleton. Therefore, the xPt/yCo3O4/3DOM Al2O3 samples possessed a hierarchically ordered macro-/mesoporous morphology. The 1.2Pt/8.8Co3O4/com-Al2O3 sample, however, exhibits a wormhole-like mesoporous structure (Fig. 3(k) and (l)), and a number of Co3O4 and/or Pt NPs are uniformly dispersed on the Al2O3 surfaces. The lattice spacings (d values) of the Co3O4 NPs in the 9.2Co3O4/3DOM Al2O3,xPt/yCo3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 samples were approximately all 0.46 nm (Fig. 3(b),(d),(f),(h) and (l)), a value that is in good agreement with the (111) plane spacing in a standard Co3O4 (JCPDS PDF #42-1467) sample. The d values of the Pt NPs in the xPt/yCo3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 were estimated to be approximately 0.23 nm (Fig. 3(d),(f),(h),(j) and (l)), which is close to that of the (111) plane of a standard Pt (JCPDS PDF #87-0644) sample. The Co3O4 NPs in the 9.2Co3O4/3DOM Al2O3 and xPt/yCo3O4/3DOM Al2O3 were 18.3 nm in size, whereas the average size of the Co3O4 NPs in the 1.2Pt/8.8Co3O4/com-Al2O3 was 15.8 nm (Table 1 and Fig. 4). The average Pt particle sizes in the 0.3Pt/8.9Co3O4/3DOM Al2O3,0.6Pt/8.9Co3O4/3DOM Al2O3,1.3Pt/8.9Co3O4/3DOM Al2O3,1.4Pt/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 were 2.3,2.4,2.6,2.5 and 3.2 nm (Table 1 and Fig. 4), respectively. It is known that the crystalline phase of alumina can influence the thermal stability of its supported catalyst. We found that the amorphous Al2O3 used in this work could be converted to crystalline γ-Al2O3 after calcination at 800 ℃ for 5 h (Fig. 5), but the ordered mesopores within the skeletons were totally destroyed. It is also well known that the surface area, porous structure extent of noble metal NP dispersion are all important factors governing the performance of a catalyst. Therefore, we fabricated meso-/macroporous Al2O3 with an amorphous phase as the support material.
Fig. 6 presents the N2 adsorption-desorption isotherms and pore size distributions of the samples. Each of the 3DOM Al2O3-supported samples displayed a type Ⅱ isotherm with an H3 hysteresis loop in the relative pressure (p/p0) range of 0.8-1.0, indicating the presence of macropores (Fig. 6(a)- (1-5)) [28]. In addition, mesopores were formed, as confirmed by the pore size distributions (Fig. 6(b)-(1-5)). These data indicated that the 3DOM Al2O3-supported samples possessed bimodal (macro-/mesoporous) structures. The N2 adsorption-desorption isotherm of the 1.2Pt/8.8Co3O4/com-Al2O3 sample exhibited a type Ⅳ isotherm with an H1 hysteresis loop, characteristic of cylindrical mesopores (Fig. 6(a)-(6)). The broad pore size distribution indicates that the 1.2Pt/8.8Co3O4/com-Al2O3 did not contain uniform mesopores (Fig. 6(b)-(6)). Table 1 summarizes the textural parameters of the as-prepared samples. The BET surface areas (94-100 m2/g) of the 3DOM Al2O3-supported samples were slightly lower than that (110.2 m2/g) of the 1.2Pt/8.8Co3O4/com-Al2O3 sample.
The XPS technique was used to investigate the surface element compositions, metal oxidation states and adsorbed oxygen species of the as-prepared samples, and Fig. 7 shows the Co 2p3/2, O 1s and Pt 4d XPS spectra of the samples. The asymmetrical Co 2p3/2 signal of each sample (Fig. 7(a)) could be deconvoluted to three components at BE values of 780.2,781.9 and 786.5 eV, which were assignable to the surface Co3+ and Co2+ species [29] and the satellite of the surface Co2+ species [30], respectively. Table 2 summarizes the surface Co3+/Co2+ molar ratios of the samples, from which it is evident that the surface Co3+/Co2+ molar ratio decreased (that is, the surface Co2+ concentration increased) after loading the Pt NPs on the 9.2Co3O4/3DOM Al2O3 surface. The surface Co3+/Co2+ molar ratio (1.95) of the 9.2Co3O4/3DOM Al2O3 was much higher than those (0.89-1.15) of the xPt/yCo3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3. The 1.3Pt/8.9Co3O4/3DOM Al2O3 sample exhibited the lowest surface Co3+/Co2+ molar ratio (0.89), and the surface Co3+/Co2+ molar ratio (0.95) of the 1.2Pt/8.8Co3O4/com-Al2O3 was lower than that (1.15) of the 0.3Pt/8.9Co3O4/3DOM Al2O3 and that (1.01) of the 0.6Pt/8.9Co3O4/3DOM Al2O3. These results demonstrate that there were more surface oxygen vacancies on the Pt and Co3 O4-loaded samples than on the 9.2Co3O4/3DOM Al2O3 [31]. Such structural defects can facilitate the adsorption and activation of gas phase oxygen molecules, thereby enhancing the catalytic performance of the materials for toluene oxidation [32].
As shown in Fig. 7(b), the asymmetrical O 1s spectrum of each sample could be deconvoluted to two components, one with BE = 530.5 eV, ascribable to surface lattice oxygen (Olatt) species, and the other at BE = 531.6 eV, attributable to surface adsorbed oxygen (Oads, such as O2-, O22- and O-) species [33]. Following the loading of Pt on the 9.2Co3O4/3DOM Al2O3 surface, the surface Oads/Olatt molar ratio increased markedly (Table 2), with the highest Oads/Olatt molar ratio (2.45) being achieved on the 1.3Pt/8.9Co3O4/3DOM Al2O3 sample. The changes in the Oads/Olatt molar ratios of the samples display a trend similar to that of the surface Co3+/Co2+ molar ratios.
Fig. 7(c) presents the Pt 4d XPS spectra of the xPt/yCo3O4/3DOM Al2O3,1.4Pt/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3. Since the Al 2p and Pt 4f lines overlap, the Pt 4d line was used in the present study. The Pt 4d spectrum of each sample could be deconvoluted into three components. The component at BE = 313.2 eV was assigned to surface metallic Pt (Pt0) species [34], while that at BE = 315.6 eV was attributed to typical surface PtO species [35], and the component at BE = 318.6 eV was higher than expected for pure PtO2 [36]. The detection of surface Ptδ+ species indicates a strong interaction between Pt and the Co3O4 NPs via the reaction Pt0 + Co3+ Ptδ+ + Co2+. The surface Ptδ+/Pt0 molar ratio (2.82) of the 1.3Pt/8.9Co3O4/3DOM Al2O3 was much higher than those (1.04-2.07) of the other supported Pt samples (Table 2). Therefore, the 1.3Pt/8.9Co3O4/3DOM Al2O3 possessed the highest Oads/Olatt and Ptδ+/Pt0 molar ratios as well as the lowest Co3+/Co2+ molar ratio. The increased Oads concentration resulting from a greater number of oxygen vacancies would be beneficial in terms of enhancing the catalytic performance of the 1.3Pt/8.9Co3O4/3DOM Al2O3 sample for toluene oxidation.
Given that the reducibility of a catalyst can play an important role in redox-based reactions, we carried out TPR experiments to investigate the reducibility of the as-prepared samples, and the resulting profiles are shown in Fig. 8(a). There were two reduction peaks for the 9.2Co3O4/3DOM Al2O3 (Fig. 8(a)-(1)). The peak centered at 495 ℃ was due to the reduction of Co3O4 to CoO, whereas the peak at 606 ℃ was attributed to the reduction of CoO to Co0 [37]. In the case of the 1.4Pt/3DOM Al2O3 (Fig. 8(a)-(5)), two weaker reduction peaks appeared at 273 and 394 ℃, ascribed to the reduction of chemically adsorbed oxygen species on the highly dispersed Pt NPs (that is, from Pt-Ox to Pt0) [38, 39, 40]. After loading of the Pt NPs, the xPt/8.9Co3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 reduction peaks (Fig. 8(a)-(2,3,4,6)) were shifted to lower temperatures, especially in the case of the 1.3Pt/8.9Co3O4/3DOM Al2O3 sample (Fig. 8(a)-(4)). This result implies a strong interaction between the Pt and the Co3O4 NPs, resulting in improved low-temperature reducibility. By quantifying the reduction peaks, one can obtain the total H2 consumption of the samples, as summarized in Table 2. It can be seen that the total H2 consumption values of the 9.2Co3O4/3DOM Al2O3,xPt/8.9Co3O4/3DOM Al2O3,1.4Pt/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 were 1.16,1.37-3.48,1.39 and 2.02 mmol/g, respectively. In addition to the highest total H2 consumption (3.48 mmol/g), the 1.3Pt/8.9Co3O4/3DOM Al2O3 also possessed the highest low-temperature H2 consumption (0.41 mmol/g), demonstrating that this sample exhibited the best low-temperature reducibility.
To better evaluate the low-temperature reducibility of the samples, we calculated the initial H2 consumption rates (the rates up to the point at which less than 25% of the oxygen was removed for the first reduction peak) [41], as shown in Fig. 8(b). The initial H2 consumption rate (equivalent to the low-temperature reducibility) increased in the order of 9.2Co3O4/3DOM Al2O3 < 0.3Pt/8.9Co3O4/3DOM Al2O3 < 0.6Pt/8.9Co3O4/3DOM Al2O3 < 1.4Pt/3DOM Al2O3 < 1.2Pt/8.8Co3O4/com-Al2O3 < 1.3Pt/8.9Co3O4/3DOM Al2O3. This trend in the low-temperature reducibility was in good agreement with the observed Oads concentration and catalytic performance data, as presented below.
A blank trial in which only quartz sand was loaded in the microreactor and the temperature was maintained below 400 ℃ was carried out using a toluene concentration of 1000 ppm, a toluene/O2 molar ratio of 1/400 and an SV of 20000 mL g-1 h-1. No significant conversion of toluene was detected under these conditions, demonstrating that there were no obvious homogeneous reactions under the adopted reaction conditions. Fig. 9 summarizes the catalytic activities of the as-prepared samples during the oxidation of toluene. The toluene conversion is seen to have increased monotonically with increasing temperature. It is preferable to use the reaction temperatures T10%,T50% and T90% (corresponding to toluene conversions of 10%,50% and 90%) to evaluate the catalytic performance of the samples, as summarized in Table 3. The 1.3Pt/8.9Co3O4/3DOM Al2O3 sample exhibited the best performance, with T10%,T50% and T90% values of 106,140 and 160 ℃ at SV = 20000 mL g-1 h-1. In addition, the Pt- and/or Co3O4-loaded samples outperformed the sample loaded with only Co3O4. The activity over the 1.3Pt/8.9Co3O4/3DOM Al2O3 was much higher than that (T10% = 140 ℃,T50% = 164 ℃ and T90% = 180 ℃) over the 1.2Pt/8.8Co3O4/com-Al2O3.
Table 4 summarizes the activities for toluene oxidation over various catalysts that have been reported in the references. By comparing the activities of these samples with that of the best-performing 1.3Pt/8.9Co3O4/3DOM Al2O3 from the present study, it is evident that the catalytic activity (T90% = 160 ℃ at 20000 mL g-1 h-1) over the 1.3Pt/8.9Co3O4/3DOM Al2O3 was superior to that (T90% = 215 ℃ at 18000 mL g-1 h-1) obtained from 1.0% Pt/γ-Al2O3, as well as that (T90% = 215 ℃ at 8400 mL g-1 h-1) over 1% Pt/Al2O3−30 wt% CeO2 and (T90% = 226 ℃ at 20000 mL g-1 h-1) over 4.6% Au/macroporous LaMnO3. The value from this work was also similar to the value (T90% = 170 ℃ at 20000 mL g-1 h-1) over 6.4% Au/3DOM La0.6Sr0.4MnO3, but inferior to that (T90% = 138 ℃ at 20000 mL g-1 h-1) over 6.5% Au/meso-Co3O4 and (T90% = 149 ℃ at 30000 mL g-1 h-1) over 1.0% Pt/ZSM-5. Compared with Au, Pt is more expensive, and a loading of 1.3% is relatively high. Thus, with regard to practical applications, it is necessary to reduce the loading of Pt in these catalysts. Doping with a base transition metal such as Co might be an effective means of reducing the noble metal loading without a significant loss in catalytic activity. In fact, we have recently found that supported Pt-Co nanocatalysts exhibit excellent performance for the catalytic oxidation of typical VOCs, which will be reported in the future.
To examine the catalytic stability of these materials, a 48-h on-stream toluene oxidation experiment was conducted using the best-performing 1.3Pt/8.9Co3O4/3DOM Al2O3 sample at 160 ℃ and SV = 20000 mL g-1 h-1, with the results shown in Fig. 10(a). In addition,Fig. 10(b) shows the catalytic activities over fresh and used 1.3Pt/8.9Co3O4/3DOM Al2O3 after toluene oxidation at SV = 20000 mL g-1 h-1 in the presence of 1000,2000 or 3000 ppm toluene for 30,9 and 9 h, respectively. It is apparent that no significant loss in catalytic activity occurred. Furthermore, the change in toluene concentration had no obvious impact on the catalytic activity of the 1.3Pt/8.9Co3O4/3DOM Al2O3. Therefore, the 1.3Pt/8.9Co3O4/3DOM Al2O3 was catalytically durable under the adopted conditions.
Fig. 11 shows the effect of moisture on the catalytic activity during toluene oxidation over the 1.3Pt/8.9Co3O4/3DOM Al2O3. A 5.0% water vapor concentration was introduced to the reaction system gas mixture feed after the catalytic activity at 160 ℃ and SV = 20000 mL g-1 h-1became stable. It can be seen from these data that the water vapor decreased the activity by approximately 5%. After the water vapor feed was cut off, the activity was restored to that of the fresh sample, indicating that deactivation of the 1.3Pt/8.9Co3O4/3DOM Al2O3 induced by water vapor addition was reversible. The negative effect of moisture on toluene oxidation is attributed to the competitive adsorption of water in place of toluene or oxygen molecules. It is worth noting that the doping of Al2O3 with CeO2 can improve the moisture tolerance of the catalyst. Recently, we determined that the introduction of water vapor to the reaction system when using the xPt/3DOM CeO2-Al2O3 catalyst has a positive effect on the catalytic activity during the oxidation of toluene.
Toluene consumption rates versus reaction temperature over the as-prepared samples are shown in Fig. 12. Toluene consumption rates were found to increase in a linear fashion with rising temperature. It is known that there are several different types of active sites (such as noble metals, transition-metal oxides and interfaces between noble metals and transition-metal oxides) in reducible metal oxide-supported noble metal catalysts, making it difficult to identify a single type of active site. Therefore, the accurate calculation of the turnover frequency can be challenging. For comparison purposes, however, we calculated the toluene oxidation rate (r) as xV0/mcat (or nM), where x is the conversion at a given temperature,V0 is the initial toluene concentration per second,mcat is the mass of catalyst, and nM (M = Pt or Co3O4) is the molar amount of M. Table 3 summarizes the toluene oxidation rates at 160 ℃ over the samples. It can be seen that the r value increased in the order of 9.2Co3O4/3DOM Al2O3 < 0.3Pt/8.9Co3O4/3DOM Al2O3 < 0.6Pt/8.9Co3O4/3DOM Al2O3 < 1.4Pt/3DOM Al2O3 < 1.2Pt/8.8Co3O4/com-Al2O3 < 1.3Pt/8.9Co3O4/3DOM Al2O3 when normalized by grams of catalyst, and in the order of 9.2Co3O4/3DOM Al2O3 < 0.3Pt/8.9Co3O4/3DOM Al2O3 < 0.6Pt/8.9Co3O4/3DOM Al2O3 < 1.2Pt/8.8Co3O4/com-Al2O3 < 1.3Pt/8.9Co3O4/3DOM Al2O3 when normalized by moles of Co3O4. However, the r value of the 1.4Pt/3DOM Al2O3 was 0.667 mmol mol-1 s-1) for Pt. This value was lower than those (0.762 to 5.557 mmol mol-1 s-1) obtained from the xPt/8.9Co3O4/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3, possibly owing to the much higher Pt loading in the former sample. Therefore, the 1.3Pt/8.9Co3O4/3DOM Al2O3 showed the highest toluene oxidation rate. In addition to the Co3O4 and Pt NP active sites, the interaction between the Pt and the Co3O4 NPs might also play an important role in enhancing the oxidation of toluene over 1.3Pt/8.9Co3O4/3DOM Al2O3. The presence of a hierarchically ordered macro-/mesoporous structure would be expected to faci litate the dispersion of the Pt and the Co3O4 NPs on the pore surfaces [45] and render these active sites more accessible to the reactant molecules, thereby improving the catalytic activity.
Fig. 13 shows the toluene consumption rate for catalysts at 160 ℃ as a function of the Oads/Olatt molar ratio in each sample. It is evident that a higher Oads concentration enhances the catalytic activity of the sample, and that the 1.3Pt/8.9Co3O4/3DOM Al2O3 exhibited the highest toluene oxidation rate and Oads/Olatt molar ratio. The toluene consumption rate (0.222 mmol g-1 s-1) over the 1.3Pt/8.9Co3O4/3DOM Al2O3 was 111,7.6 and 2.6 times the values obtained over the 9.2Co3O4/3DOM Al2O3,1.4Pt/3DOM Al2O3 and 1.2Pt/8.8Co3O4/com-Al2O3 (0.002,0.029 and 0.086 mmol g-1 s-1), respectively. This variation in activity was due to the interactions between the Pt and the Co3O4 NPs and the ordered macro-/mesoporous structures. Similar interactions between precious metals (such as Pt, Pd, Rh, Ru and Ag) and a CeO2 support were also observed by Acerbi et al. [46]. The interaction between the Pt and the Co3O4 NPs could possibly facilitate the activation of O2 molecules to generate active Oads species and thus enhance the catalytic activity of the 1.3Pt/8.9Co3O4/3DOM Al2O3. It is well known that the reducibility of metal oxides is associated with the formation of oxygen vacancies. Kleitz et al. [47] found that a metal o xide could induce surface defects more readily if it was more reducible, and thus could produce a higher Oads concentration. Based on the characterization and activity evaluations, we conclude that the excellent catalytic performance of the 1.3Pt/8.9Co3O4/3DOM Al2O3 results from highly dispersed Pt NPs, an elevated Oads concentration, good low-temperature reducibility, strong interactions between the Pt and the Co3O4 NPs, and a hierarchically ordered macro-/mesoporous structure.
It has been generally believed that the oxidation of VOCs over transition metal oxides obeys first-order reaction kinetics with regard to VOC concentration and zero-order kinetics with regard to oxygen concentration [48]. Therefore, it is reasonable to assume that toluene oxidation in the presence of excess oxygen (toluene/O2 molar ratio = 1/400) should follow a first-order reaction mechanism with respect to the VOC concentration (c), such that r = -kc = (-Aexp(-Ea/RT))c, where r,k,A and Ea are the reaction rate (mol/s), rate constant (s-1), pre-exponential factor and apparent activation energy (kJ/mol), respectively. Fig. 14 presents the Arrhenius plots for toluene oxidation over the as-prepared samples, and the calculated Ea values are summarized in Table 3. It can be seen that the Ea decreased in the order of 9.2Co3O4/3DOM Al2O3 > 0.3Pt/8.9Co3O4/3DOM Al2O3 > 0.6Pt/8.9Co3O4/3DOM Al2O3 > 1.4Pt/3DOM Al2O3 > 1.2Pt/8.8Co3O4/com-Al2O3 > 1.3Pt/8.9Co3O4/3DOM Al2O3. The Ea (42.6 kJ/mol) over the best-performing 1.3Pt/8.9Co3O4/3DOM Al2O3 catalyst was much lower than the values (120-144 kJ/mol) obtained from CuO/Al2O3 and MnOx/Al2O3 [49] as well as those (54−76 kJ/mol) over Au/3DOM Mn2O3 [50], but close to the values (45-55 kJ/mol) reported for Au/meso-Co3O4 [44]. These results indicate that toluene oxidation proceeds more readily over the 1.3Pt/8.9Co3O4/3DOM Al2O3 sample.
In this work,xPt/yCo3O4/3DOM Al2O3 samples were prepared via PMMA-templating, incipient wetness impregnation and PVA-protected reduction. The xPt/yCo3O4/3DOM Al2O3 samples displayed a hierarchical 3DOM architecture (with macropore diameters from 180 to 200 nm and mesopore diameters from 4 to 6 nm) and surface areas in the range of 94 to 102 m2/g. Co3O4 NPs that were 18.3 nm in size and Pt NPs with sizes from 2.3 to 2.5 nm were uniformly dispersed on the surfaces of the 3DOM Al2O3. Among the xPt/yCo3O4/3DOM Al2O3 catalysts, the 1.3Pt/8.9Co3O4/3DOM Al2O3 had the highest Oads concentration and the best low-temperature reducibility, and thus exhibited the highest catalytic activity (T90% = 160 ℃ at SV = 20000 mL g-1 h-1) for toluene oxidation. Based on the activity data and characterization results, we conclude that the superior performance of this material can be attributed to well dispersed Pt NPs, an elevated Oads concentration, superior reducibility at low temperatures and significant Pt to Co3O4 NP interactions. The hierarchically ordered macro-/mesoporous structure of this material is also believed to have been partly responsible for the excellent catalytic abilities of the 1.3Pt/8.9Co3O4/3DOM Al2O3.
2016,
Vol. 37 Issue (6): 934-946 
PDF (1926 KB)
2016,
Vol. 37 Issue (6): 934-946 PDF (1926 KB)