In the past decades, CO and volatile organic compounds (VOCs), generally emitted from automotive devices and various industrial processes, have been the subject of strong monitoring and legislation. In 1999, the European Union legislation on the VOS Solvent Emissions Directive (1999/13/EC) was launched.
The production of formaldehyde by selective methanol oxidation on oxide catalysts leads to toxic gas emissions (formaldehyde, CO, dimethyl ether (DME), and methanol), the latter being strictly recorded and controlled [1, 2]. Moreover, in relation with the normative documents on ecology and environment protection, the complete removal of the toxic products emitted during the process is of significant importance [3, 4]. The considerable efforts to protect the environment and modern clinical trials on the negative effects of certain VOCs on human health, including CO, have led to the application of increasingly stringent preventive measures to reduce their emission [5, 6]. Catalytic incineration has been reported as one of the most appropriate methods for their elimination. However, the development of highly effective catalysts remains under intense investigation. Supported noble metals (Pt, Pd, Rh) [7, 8] or metal oxides (Cu, Cr, Mn, Co) [9, 10] are typical catalysts for such an application. Currently, supported Au catalysts are one of the major topics of interest to the catalytic community because of their excellent activity and selectivity for many reactions. Very recently, comprehensive reviews on the use of supported gold catalysts in the catalytic deep oxidation of VOCs were published by Scire et al. [11] and Barakat et al. [12]. The performance of gold catalysts for VOCs deep oxidation depends on many factors, namely, the nature and properties of the support, loading of gold, size, shape, and electronic state of gold nanoparticles, preparation method and pretreatment conditions of catalysts, and nature and concentration of the organic molecule to be combusted. Among all the above cited factors, the choice of support plays a fundamental role. The support helps to regulate the amount of gold anchored onto the surface as well as the size and shape of the gold particles. Moreover, when reducible oxides are used as support, the latter takes part in the reaction pathway. In this case, the resulting anion vacancies (surface lattice oxygen) close to the gold particles serve as the sites of oxygen adsorption and activation.
The low temperature oxidation of CH3OH, HCHO, and HCOOH over gold supported on reducible oxides was reported by Haruta et al. [13]. As reported, a mixed catalyst system consisting of gold coupled with cobalt oxide supported over CeO2/TiO2 exhibited better performance toward the complete oxidation of methanol when compared with a Pt catalyst [14]. Wang et al. [15] demonstrated that methanol oxidation over gold/iron oxide aerogel catalysts resulted in diverse products such as DME, formaldehyde, methyl formate, and CO2. Furthermore, the CO2 production increased remarkably with increasing gold contents. Bonelli et al. [16] investigated methanol oxidation over FeOx and Au/FeOx/ceria catalysts and observed a strong positive effect on both the activity and reducibility of the catalysts after Au addition.
CeO2 alone or combined with other elements in mixed oxides or doped oxides (CoOx, MnOx, CuO, ZnO, Fe2O3, TiO2, Al2O3, ZrO2) is recognized as an active catalyst for the total oxidation of different types of VOC molecules [17, 18, 19, 20, 21]. Ceria is a very attractive support material with unique catalytic properties, including the ability to enhance precious metal dispersion and catalytic activity at the interfacial metal-support sites, and rate of formation and elimination of oxygen vacancy defects. Tabakova et al. [22] studied the catalytic performance of CO and CH3OH oxidation over Au catalysts supported on ceria doped with different metal oxides (i.e., Fe, Mn, and Co), and observed a strong influence of the nature of the dopant.
Nanostructured mesoporous materials with novel pore systems and properties have attracted considerable attention because of their remarkably large surface area and narrow pore size distribution that make them ideal candidates for catalyst systems. Studies focusing on the importance of the properties of the support were conducted, whereby nanostructured mesoporous oxides were used as supports for gold. The first investigation was carried out by Idakiev et al. [23], involving the oxidation of benzene over Au/V2O5 supported on mesoporous TiO2 and ZrO2. Gold catalysts supported on mesoporous TiO2, ZnO, or Al2O3 were assessed toward the oxidation of propene; Au/TiO2 was the best performing catalyst system [24]. In another study Zhang et al. [25] reported the superior catalytic activity of Au/mesoporous ZrO2 (surface area of 447 m2 g−1) relative to that of conventional Au/ZrO2 toward the combustion of HCHO. Ying et al. [26] studied gold catalysts supported on mesoporous ceria prepared using MCM-48 as a hard template. The catalyst system displayed higher stability and activity toward benzene oxidation when compared with the corresponding catalyst prepared using DP method. Gennequin et al. [27] reported efficient propene oxidation to CO2 over Au/CexTi1−xO2 (x = 0-0.3) catalysts in the range of 200-400 °C. Higher catalytic activities were obtained upon addition of a small amount of cerium to TiO2; the highest propene conversion was achieved using Au/Ce0.3Ti0.7O2 catalyst. The same authors, who investigated the catalytic performance of gold deposited on Ce-Ti mixed oxides (including Ce-Ti-Zr) prepared by sol-gel, concluded that mesoporous materials employed as supports of gold-based catalysts have important and interesting effects on the oxidation activity of VOCs [28]. It is believed that the application of such advanced materials, modified by ceria, as gold nano-catalyst supports, would be an efficient approach to enhance catalytic activities in important reactions for environmental protection.
The aim of the present study was to examine the role of ceria as a modifying additive to mesoporous oxides (Ce/TiO2 and Ce/ZrO2) and as a support for Au catalysts in the catalytic abatement of air pollutants, i.e., CO, CH3OH, and DME. These VOCs were selected because their presence in emission gases from the formaldehyde production industry represents an ecological problem.
The mesoporous TiO2 and ZrO2 (MTi and MZr, respectively) were obtained using a non-ionic surfactant as a templating agent, as described by Idakiev et al. [23].
MTi was synthesized using a surfactant (C13(EO)6- polyoxyethylene(6) tridecylether) templating method via a neutral C13(EO)6-Ti(OC3H7)4 assembly pathway. The hydrothermal treatment was conducted at 60 °C for 48 h. The template was completely removed following ethanol extraction for 48 h using a Soxhlet apparatus. MTi was dried under vacuum at 80 °C, then calcined in air at 400 °C for 2 h.
MZr was prepared using the same synthesis method as that described above using the same non-ionic surfactant and zirconium n-propoxide (Zr(OC3H7)4) as Zr source. The surfactant/zirconia molar ratio was 1.5. The hydrothermal treatment was conducted at 80 °C for 24 h. MZr was dried and calcined under the same conditions stated above.
Ceria was loaded onto the mesoporous oxides using a DP method [29]. Typically, ceria was deposited on the mesoporous oxides by precipitation of Ce(NO3)3·6H2O with Na2CO3 at 60 °C and pH 9.0. The content of ceria in the supports was 20 wt%. The samples were denoted as CeMTi and CeMZr.
Varying contents of gold (1-3 wt%) were loaded on the mesoporous oxides (denoted as Au/MTi and Au/MZr) and ceria-modified supports (denoted as AuCeMTi and AuCeMZr) using the DP method. Typically, gold hydroxide was supported on the metal oxide supports, suspended in water, via chemical reaction between HAuCl4·3H2O and Na2CO3 in aqueous solution [30]. The measured gold loading in the catalysts is shown in Table 1.
The X-ray diffraction (XRD) patterns were obtained from a Philips PW 170 diffractometer using Cu Kα (λ = 1.54178 Å) radiation.
The transmission electron microscopy (TEM) images were taken using a Philips Tecnai-10 microscope operating at 100 kV. Prior to analysis, the sample powders were embedded in epoxy resin and sectioned with an ultramicrotome. The sections were then placed on copper grids previously coated with carbon to improve stability and reduce the accumulation of charges. High-resolution transmission electron microscopy (HRTEM) analysis was performed on a JEOL JEM-3010 microscope operating at 300 kV. For the sample preparation, the powdered samples were ultrasonically dispersed in isopropyl alcohol and the obtained suspensions were deposited on a holey carbon film supported a copper grid.
The N2 adsorption-desorption isotherms were measured at -196 °C within a relative pressure (p/p0) range of 0.01-0.995 using a Micromeritics TriStar 3000 volumetric adsorption analyzer. The pore diameter and pore size distribution were determined by the BJH method using the adsorption branch of the isotherms [31]. The specific surface areas were determined by the BET method [32].
Temperature-programmed reduction (TPR) of the samples was carried out in the measurement cell of a differential scanning calorimeter (DSC-111, SETARAM) that was connected to a gas chromatography (GC) instrument [33].
The catalytic activity of the samples toward CO, CH3OH, and DME oxidation was assessed using a continuous flow setup, consisting of a fixed-bed stainless steel reactor, at atmospheric pressure within a temperature range of 20-360 °C. The GHSV (gas hourly space velocity) was 25000 h−1, the inlet CO, CH3OH, and DME concentration was 2.0% balanced with air (0.4 cm3 catalyst, particle size 0.3-0.6 mm). The reactant and product gases were analyzed using a HP 5890 Series II gas chromatograph equipped with flame ionization and thermal conductivity detectors and Porapak Q and MS-5A columns.
The XRD patterns of the prepared mesoporous oxides, ceria-modified mesoporous supports, and gold catalysts are shown in Fig. 1. The X-ray diffractograms of the samples based on mesoporous titania (MTi), calcined at 400 °C, are shown in Fig. 1(a), whereas Fig. 1(b) shows XRD patterns of the prepared samples based on mesoporous zirconia (MZr).
The MTi support showed diffraction peaks characteristic of anatase. Significant differences after the deposition of ceria additive were observed. The presence of ceria in the sample led to the formation of a highly defective anatase crystal structure (CeMTi). Only a weak diffraction peak at 2θ = 28.5° that was ascribed to the most intense fluorite oxide-type diffraction pattern of CeO2 was visible. Other diffraction peaks characteristic of the cubic structure of ceria (2θ = 47.48° and 56.33°) were not discernible because of their similar position to that of diffraction peaks corresponding to titania. The calculated average size of MTi was 9.7 nm. In contrast, modification with ceria generated particles with a reduced average size of 6.1 nm. The results revealed the beneficial role of ceria additive in decreasing the degree of crystallinity of MTi and its particle size. No significant differences following the deposition of gold were detected. The characteristic peaks associated with gold at 2θ = 38.2° and 44.4° were difficult to detect in the gold-containing samples because of the very fine dispersion of the Au nanoparticles on the surface of the mesoporous TiO2 and ceria-modified mesoporous TiO2 supports.
After calcination of MZr at 400 °C, peaks corresponding to the tetragonal structure of crystalline zirconia were observed at 2q = 30°, 34.5°, and 50° [34]. A highly defective tetragonal structure of ZrO2 was obtained following deposition of ceria additive. The typical diffraction patterns of ZrO2 were broader and less prominent. The absence of characteristic peaks of gold was attributed to the fine dispersion of the gold nanoparticles on the surface of CeMZr. The mean crystallite size of MZr (6.8 nm) and CeMZr (5.6 nm) was determined using Scherrer’s equation. Thus, MZr crystallites were considerably larger than those of the binary oxides CexMZr1−xO2 (x = 0.2), indicating that the incorporation of Ce4+ ions into mesoporous zirconia inhibited the growth of the tetragonal crystallites.
The N2 adsorption-desorption isotherms and pore size distribution of the supports MTi and CeMTi and gold-based catalyst AuCeMTi are depicted in Fig. 2.
The textural properties of the supports and catalyst samples are listed in Table 1. All samples displayed type IV isotherms, characteristic of mesoporous materials according to the Brunauer-Deming-Deming-Teller (BDDT) classification [35]. As observed, the adsorption volume of N2 increased sharply in the p/p0 range of 0.60-0.90, relevant to capillary condensation and indicating good homogeneity of the samples. The pore size distribution of the samples, as modeled by the BJH method, was relatively narrow, confirming the good quality of the samples. The specific surface area (SBET) of as-prepared MTi (prior to calcination) was 204 m2/g and decreased to 125 m2/g following calcination at 400 °C, whereas that of CeMTi decreased from 165 to 103 m2/g. Ceria deposition on mesoporous titania led to significantly lower surface areas and pore volumes (Table 1), likely because of the presence of ceria additive in the pores. The decrease in the surface area upon calcination is related to the crystallization of the walls separating the mesopores as previously reported [36]. In contrast, it is interesting to note that the values of SBET of the gold-containing catalysts after calcination at 400 °C in comparison with those of MTi and CeMTi were higher. Thus, it can be concluded that the deposition of gold enhances the thermal stability of the supports. The pore diameter of the gold-loaded catalyst on MTi support was lower after calcination at 400 °C when compared with that of MTi calcined at 400 °C. In contrast, both the CeMTi support and gold-loaded catalyst supported on CeMTi have comparable pore diameters (Table 1). The shape of the isotherms and the pore sizes did not change significantly, indicating that the mesoporous structure of the material was retained following introduction of ceria. Likewise, the pore size distribution was relatively narrow, confirming the good quality of the sa mples.
Figure 3 shows the N2 adsorption-desorption isotherms and the corresponding pore size distributions of the supports MZr and CeMZr and gold-based catalyst AuCeMZr. The isotherms displayed a linear region in the p/p0 range of 0.1-0.4 before reaching a plateau at p/p0 = 0.85, indicative of mesoporous materials, with pore sizes of ~2.0 nm [37].
At relative pressures above 0.90, the adsorbed volume of N2 increased significantly rather than remaining stable owing to saturation. This finding suggests that the samples contain an appreciable amount of secondary mesoporosity or macroporosity. The additional mesoporosity and/or macroporosity is expected to enable efficient transport of reagents to the framework reaction centers, thereby favoring the catalytic activity. The surface area of as-prepared MZr was 613 m2/g and decreased to 142 m2/g after calcination at 400 °C, whereas the pore diameter increased from 2.1 to 2.7 nm (Table 1). As observed, the specific surface area of the sample was significantly influenced by its composition. Following ceria deposition on MZr, the surface area decreased significantly. Likewise, the pore volume decreased (Table 1). This phenomenon was also observed in the prepared ceria-modified mesoporous TiO2 discussed above. Likewise, the reduction in the specific surface area and pore volume in the ceria-modified MZr support could be attributed to the presence of ceria additive in the pores. The decrease in the surface area upon calcination was also due to the crystallization of the walls separating the mesopores, as previously reported [36]. The differences in the surface areas and pore volumes of the support CeMZr and corresponding supported gold catalysts were insignificant.
TEM analysis of the gold catalyst prepared on the ceria-modified mesoporous TiO2 support (Fig. 4(a)) revealed that the gold nanoparticles were homogeneously dispersed. The gold loading and particle size are given in Table 1. As observed from the TEM images, the mesoporous TiO2 support and corresponding gold catalyst featured a disordered structure with numerous wormhole-like channels [38].
The TEM image of the gold catalyst prepared on the ceria-modified mesoporous ZrO2 support (Fig. 4(b)) shows a disordered structure and aggregation of the nanoparticles. HRTEM analyses revealed that the gold particles with sizes of ~5 nm were homogeneously dispersed in the mixed metal oxide support. As observed from the selected area electron diffraction pattern of AuCeMZr, a strong electron diffraction pattern was obtained that was indicative of the good dispersion of Au on AuCeMZr.
Figure 5 shows the TPR profiles of the mesoporous oxide supports (MTi, MZr), ceria-modified mesoporous oxide supports (CeMTi, CeMZr), gold-based catalysts (AuCeMTi, AuCeMZr), and ceria for reference.
As observed in Fig. 5(a), MTi displayed a single peak above 500 °C corresponding to the reduction of TiO2. Modification with ceria considerably influenced the reducibility of mesoporous TiO2. The H2-TPR profile of ceria featured two major peaks, one at a lower temperature (~500 °C), corresponding to the reduction of surface oxygen, and another peak at a higher temperature (~800 °C), corresponding to the removal of bulk oxygen from the ceria structure as reported by Yao et al. [39]. A significant increase in the intensity of the peak at 470 °C ascribed to ceria surface oxygen reduction was observed in the ceria-modified mesoporous titania. The higher hydrogen consumption could be related to the synergistic interaction between ceria and mesoporous TiO2. Hence, it can be deduced that the presence of gold facilitates the reducibility of the surface oxygen of ceria-modified mesoporous TiO2. A strong and sharp low-temperature peak at Tmax 112 °C was observed in the AuCeMTi. A recent TPR study on Au/ceria catalysts prepared by the DP method has shown that hydrogen consumption at the low temperature is related to two processes, namely, the reduction of oxygen species on the nano-sized gold particles and surface reduction of ceria [40].
As observed, mesoporous zirconia calcined at 400 °C only displayed a high-temperature peak above 600 °C, corresponding to the reduction of ZrO2 reduction (Fig. 5(b)). Likewise, ceria modification heavily influenced the reducibility of mesoporous ZrO2. Differences were observed in both the intensity and shape of the TPR peaks of mesoporous ZrO2 and ceria-modified mesoporous ZrO2. A significant increase in the intensity of the peak at 480 °C, corresponding to ceria surface oxygen reduction, was detected in the TPR profile of CeMZr. The higher hydrogen consumption over CeMZr at 580 °C could be related to the synergistic interaction between ceria and mesoporous zirconia, in agreement with the results of the other characterization methods employed and Zr4+-to-Zr3+ reduction. In the presence of gold, the reducibility of the surface oxygen of ceria-modified mesoporous ZrO2 was significantly enhanced. A strong and sharp low-temperature peak at Tmax 115 °C was observed for AuCeMZr.
The prepared catalyst samples were assessed toward the oxidation of CO, methanol, and DME, and the results are shown in Figs. 6-9. The catalytic measurements demonstrated the strong influence of the nature of the mesoporous oxide, ceria modification, and gold loading on the catalytic activity of the catalysts.
The temperature-dependence of CO oxidation degree of the mesoporous supports (MTi, MZr), ceria-modified mesoporous oxide supports (CeMTi, CeMZr), and gold-based catalysts is presented in Fig. 6.
As observed, AuCeMZr showed superior activity over the other catalysts. The catalytic activity toward the oxidation of CO decreased as follows: AuCeMTi > AuMTi > CeMTi > MTi (Fig. 6(a)) and AuCeMZr > AuMZr > CeMZr > MZr (Fig. 6(b)) for the samples prepared on the MTi and MZr supports, respectively. Comparison of the CO oxidation activity data showed that the gold catalysts prepared on the MZr support displayed better catalytic behavior than those prepared on MTi. The gold catalyst supported on ceria-modified mesoporous zirconia displayed a considerably high catalytic activity (~100% conversion of CO at 10 °C), whereas AuCeMTi achieved 100% conversion at 60 °C. Because of the varying conditions (e.g., CO content in reaction mixture, space velocity) employed in the literature to assess the catalytic performance as well as the different gold loading in diverse studied catalysts, calculating the specific rates allows relevant comparison of the current results with those reported in the literature. The specific reaction rates for CO oxidation, expressed as molCO h-1 gAu-1, are shown in Fig. 7.
Comparison of the CO, CH3OH, and DME oxidation activity data revealed that the gold-based samples prepared on the MTi support featured a higher catalytic activity than those prepared on MZr. However, the ceria-modified mesoporous oxides are of much interest as potential supports for gold-based catalysts. The resulting supported catalysts featured improved catalytic performance because of (1) the beneficial role of ceria additive in decreasing the degree of crystallinity of the mesoporous oxide and its particle size, (2) the strong effect on the reducibility of the support owing to the interaction between the ceria additive and mesoporous oxide, and (3) the high degree of synergistic interaction between ceria and mesoporous oxide. The high activity could be related to the high stability of the gold dispersion and the higher number of active sites located at the Au/ceria-modified mesoporous oxide interface.
The catalysts displayed similar catalytic behavior towards the oxidation of methanol and DME. In general, the gold catalyst deposited on ceria-modified mesoporous oxide displayed the highest catalytic activity, followed by the gold catalyst prepared on mesoporous titanium TiO2 (Figs. 8 and 9). More specifically, regarding the oxidation of methanol, the catalysts AuCeMTi and AuMTi possessed similar catalytic activities (nearly 100% conversion was attained at ~120 °C), whereas AuCeMZr showed a significantly higher activity than AuMZr (100% conversion at 60 and 200 °C was attained for AuCeMZr and AuMZr, respectively). Additionally, the mesoporous oxides modified with cerium oxide exhibited higher activity than the pure mesoporous oxides at comparatively higher temperatures.
Comparison of the DME oxidation activity data of all the investigated samples (Fig. 9) confirmed similar catalytic activity trends to those previously discussed for the oxidation of methanol. However, the temperatures were significantly higher and the conversion values of DME were lower.
The catalytic performance of CO, CH3OH, and DME oxidation of gold catalysts supported on mesoporous and ceria-modified mesoporous titania and zirconia was studied. The gold catalyst supported on ceria-modified mesoporous zirconia displayed superior catalytic activity, achieving 100% conversion of CO at 10 °C and 100% conversion of methanol at 60 °C. The catalytic activity toward the oxidation of CO decreased in the order of AuCeMTi > AuMTi > CeMTi > MTi for the samples prepared on the MTi support and AuCeMZr > AuMZr > CeMZr > MZr for the samples prepared on the MZr support. The XRD results revealed the beneficial role of ceria additive in decreasing the degree of crystallinity and particle size of the mesoporous oxides. XRD and HRTEM analyses revealed differences in the average size and distribution of the gold particles. Taking into account the results from the H2-TPR analysis, the following can be concluded: (1) the ceria additive interacts with the mesoporous oxides and strongly influences the reducibility of the supports; (2) gold loading promotes the reducibility of the ceria-modified mesoporous oxides. The synergy between gold and ceria additive significant enhanced the reducibility and capability for oxygen activation that resulted in improved oxidation activities. The present results revealed the potential applicability of the current catalysts for the abatement of CO and VOCs.
Acknowledgments The Bulgarian authors gratefully acknowledge the financial support by the Bulgarian National Science Fund (Project FNI T02/4). V.I. and T.T. thank the National Science Fund of Bulgaria for a financial support through Project E-01/07 2012.
2015, Vol. 36 Issue (4): 579-587
PDF (683 KB)
2015, Vol. 36 Issue (4): 579-587
PDF (683 KB)