Catalytic Properties of CeO₂-Supported LaMnO₃ for NO Oxidation

Abstract

CeO₂-supported LaMnO₃ perovskite oxides were prepared to study their catalytic properties in the oxidation of NO to NO₂. To prepare the catalyst and investigate the interaction between LaMnO₃ and CeO₂, two deposition methods were used. Extended X-ray absorption fine structure studies confirmed that perovskite oxide phases were formed on the CeO₂ support. Moreover, X-ray photoelectron spectroscopy and temperature-programmed reduction with H₂ studies revealed that the reduction temperatures for perovskite oxides and CeO₂ support decreased by the deposition followed by calcination at 650 °C, and that the interaction between the LaMnO₃ and CeO₂ support can be controlled by changing the preparation method. The LaMnO₃/CeO₂ catalyst in which LaMnO₃ was highly dispersed on CeO₂ exhibited higher NO oxidation activity than either LaMnO₃ or CeO₂. The thermal stability of the LaMnO₃/CeO₂ catalyst was compared with that of alumina-supported LaMnO₃ catalysts over 850–1050 °C.

Graphical Abstract

1 Introduction

Automobile emission control technologies are attracting considerable attention due to the severe tightening of regulations and increasing vehicle production volumes. Catalytic NO oxidation is a critical process for the purification of diesel exhaust because the resulting NO₂ species is a key substance that promotes the oxidation of diesel soot and facilitates selective catalytic reduction and NOx storage under lean burn conditions. Supported Pt catalysts exhibit high activity in the oxidation of NO to NO₂ [1–3]. However, it is essential that the amount of noble metals used in the catalysts be reduced or replaced by less expensive materials.

Perovskite-type mixed oxides, represented by a ABO₃-type formula, have been widely used as catalysts for oxidation processes [4–6]. Perovskite oxides are good substitute candidates for noble metal catalysts [7]. Perovskite oxides containing Mn [8–10] and Co [8, 11, 12] at the B site and La at the A site have been frequently used for the catalytic oxidation of NO to NO₂. In perovskite oxides, the A-site cations are partially substituted by Sr [13, 14] and Ce [15] and the B-site cations are partially substituted by other transition metals [16, 17]. Thus, one method for enhancing the catalytic activity of perovskite oxides is to substitute constituent metals and cations. Another method is to deposit them on supporting materials with a high surface area. To date, ZrO₂ [18, 19], Al₂O₃ [20], CeO₂ [21], and SiO₂-base mesoporous materials [22, 23] have been used to support perovskite deposition. The supported perovskite oxides reportedly exhibit high catalytic activity and thermal stability, which effectively offset the drawback of perovskite oxides—their low surface area.

We recently reported the selective deposition of the perovskite oxides LaFeO₃ and LaMnO₃ inside or outside the pore of γ-Al₂O₃ [24, 25]. These supported perovskite oxide catalysts exhibited higher activity and thermal stability than an alumina-supported Pt catalyst when the perovskite phases were deposited inside the alumina pore. These findings inspired us to investigate the effect of supporting materials on the catalytic properties of supported perovskite oxides. Here, we investigated the effect of depositing LaMnO₃ perovskite oxides onto a CeO₂ support on the catalytic properties of the oxidation of NO to NO₂. Because CeO₂ itself is active with respect to oxidation reactions, we anticipated that the combination of CeO₂ with perovskite oxide would improve the catalytic properties of the supported catalysts. We also investigated the effect of the deposition method on their catalytic properties and thermal stabilities, which were important factors when the catalysts were put into practical use for automobile emission control.

2 Experimental

2.1 Materials

The catalyst support CeO₂ (JRC-CEO-3; S_BET = 72 m² g⁻¹) was obtained from the Catalysis Society of Japan and was used after calcination at 650 °C for 5 h in air. γ-Al₂O₃ (JRC-ALO-8) was also used as a supporting material after the same treatment.

2.2 Preparation of LaMnO₃/CeO₂ Catalysts

Bulk LaMnO₃ perovskite oxides were prepared using the hydrolysis-precipitation method described in a previous paper [26]. The aqueous solution containing La(NO₃)₃ (Wako Pure Chem) and Mn(NO₃)₂·6H₂O (Wako Pure Chem) were slowly added to aqueous ammonia while vigorously stirring. The resulting precipitates were calcined at 650 °C for 5 h in air. The LaMnO₃ catalyst thus prepared has orthorhombic structure with the tolerance factor of 0.954. For the preparation of physical mixture of LaMnO₃ and CeO₂, 0.20 g of LaMnO₃ and 0.80 g of CeO₂ were ground with a mortar and pestle. The sample was denoted by LaMnO₃-MM.

To prepare CeO₂-supported LaMnO₃ (LaMnO₃/CeO₂) catalysts by the dry impregnation (DI) method, CeO₂ was impregnated with a mixed aqueous solution of metal nitrates in a quantity equal to 50–75 % pore volume of the CeO₂, followed by drying at 100 °C. The impregnation process was repeated until LaMnO₃ loading was increased to 20 wt%. The obtained samples were then calcined at 650–1050 °C for 5 h in air. The LaMnO₃/CeO₂ catalyst prepared by the DI method was denoted as LaMnO₃/CeO₂-DI.

LaMnO₃/CeO₂ catalysts were also prepared by a precipitation–deposition (PD) method. Hydroxide precursors of perovskite oxide were obtained by the hydrolysis–precipitation method described above. They were then mixed well in distilled water (300 mL) with CeO₂, followed by ultrasonication. Next, they were evaporated to dryness under vigorous stirring followed by grinding and calcination at 650 °C for 5 h in air. In this process, the loading level of perovskite phase after calcination depends on the water content of the hydroxide precursor. For the determination of LaMnO₃ loading on the CeO₂ support, therefore, the water content of the hydroxide precursor was measured by thermogravimetry (Shimadzu DTG-60) prior to the impregnation of the hydroxide precursor. The LaMnO₃/CeO₂ catalyst prepared by the PD method was denoted as LaMnO₃/CeO₂-PD.

2.3 Catalyst Characterization

The crystal structure of the catalysts was examined by X-ray diffraction (Rigaku Ultima IV) using CuKα radiation. The catalyst surface area was determined using Brunauer–Emmett–Teller (BET) plots obtained from the N₂ adsorption isotherms at 77 K (Quantachrome NOVA2000).

La K-edge extended X-ray absorption fine structure (EXAFS) spectra were taken with the photon factory advanced ring at the High Energy Accelerator Research Organization (NW-10A), with the storage ring operating at an energy of 6.5 GeV. Fourier transform-EXAFS spectra were obtained from k³-weighted EXAFS data [k³χ(k)] at 3.0–12.0 Å⁻¹. The coordination numbers (CNs), bond distances (R), Debye–Waller factor (σ²), and energy shift (∆E) were obtained using the theoretical backscattering amplitude and phase shift functions calculated by the program FEFF8 [27].

X-ray photoelectron spectroscopy (XPS) spectra were recorded using a Kratos AXIS-165 spectrometer with an Al Kα source. The binding energies were corrected using the value of 284.8 eV as an internal standard for the C 1s peak of the carbon species on the catalyst samples.

Temperature-programmed reduction with H₂ (H₂-TPR) was conducted with a BELCAT-30 catalyst analyzer (BEL JAPAN, Inc). The sample (0.050 g) was pretreated in an O₂ flow for 2 h at 823 K. In the H₂-TPR measurements, catalyst samples were heated at 10 °C/min.

2.4 Catalytic Activity Measurement

The catalytic oxidation of NO was performed using a fixed bed flow reaction system. Catalyst samples meshed at 250–750 µm were placed in the U-shaped glass reactor, which was connected to the system. The reaction gas, with a NO 500 ppm-O₂ 5 %-N₂ balance composition, was fed to the reactor at 200–500 °C. In a typical reaction, the catalyst weighed 0.10 g and the gas flow was 150 mL/min (W/F = 0.04 g s mL⁻¹). The concentrations of NO and NO₂ were determined using a NOx analyzer (SHIMAZU NOA-7000). Prior to the reaction, the catalyst was heated in an O₂ flow at 550 °C. The oxidation of NO to NO₂ consists of both forward reaction (NO + 1/2O₂ → NO₂) and backward reaction (NO₂ → NO + 1/2O₂). Therefore, the steady-state activity was dominated by reaction equilibrium. The equilibrium curve for conversion of NO to NO₂ was calculated based on the van’t Hoff Eqs. (1) and (2), where K_P, ∆H, R, T, ${{\text{p}}_{\text{NO}}}$, ${{\text{p}}_{\text{N}{{\text{O}}_{2}}}}$, and ${{\text{p}}_{{{\text{O}}_{2}}}}$were equilibrium constant expressed in terms of partial pressures of gases, the enthalpy of reaction, gas constant, reaction temperature, partial pressures of NO, NO₂, and O₂.

$$\frac{\text{d}\ln {{\text{K}}_{\text{P}}}}{\text{dT}}=\frac{\text{ }\!\!\Delta\!\!\text{ H(T)}}{\text{R}{{\text{T}}^{\text{2}}}}$$

(1)

$${{\text{K}}_{\text{P}}}=\frac{{{\text{p}}_{\text{N}{{\text{O}}_{\text{2}}}}}}{{{\text{p}}_{\text{NO}}}\text{ }{{\text{p}}_{{{\text{O}}_{\text{2}}}}}^{1/2}}$$

(2)

The reaction rate was obtained under conditions in which the NO conversion was linear with respect to the ratio of catalyst weight to gas flow rate.

3 Results and Discussion

3.1 Structure of Catalysts

Figure 1 shows the XRD patterns of the CeO₂, LaMnO₃, and LaMnO₃/CeO₂ catalysts. The patterns of the CeO₂ support and LaMnO₃/CeO₂ combined were almost the same as those for the CeO₂ support alone, indicating that the structures of the CeO₂ support were unchanged after deposition and the post-heating treatment. The peaks due to the perovskite were detected for the physical mixture of LaMnO₃ and CeO₂ (LaMnO₃-MM) and the LaMnO₃/CeO₂ catalysts prepared by the PD. In contrast, perovskite phase was hardly detected for the LaMnO₃/CeO₂ catalyst prepared by the DI method, indicating that the LaMnO₃ perovskite phases were highly dispersed on the CeO₂ support.

Fig. 1

XRD patterns of LaMnO₃ and LaMnO₃/CeO₂ catalysts prepared by different methods

The BET surface areas of the LaMnO₃/CeO₂ catalysts are listed in Table 1. The surface area of LaMnO₃/CeO₂ catalysts was lower than that of the CeO₂ support when the catalyst was prepared by the DI method, although the surface area of LaMnO₃/CeO₂ catalysts was much higher than that of the unsupported LaMnO₃ catalysts. It is noted that the surface area of the catalyst prepared by the PD method was nearly comparable to that of LaMnO₃/CeO₂-MM, whereas the LaMnO₃/CeO₂-DI catalysts have a much smaller surface area than those prepared by the PD method. This implies that the DI of perovskite oxides in the pores of the CeO₂ support promotes pore blockage during the calcination process.

Table 1 BET surface area and XPS analysis data of LaMnO₃ and LaMnO₃/CeO₂ catalysts

We studied the LaMnO₃/CeO₂ catalysts using EXAFS to investigate the structure of the mixed oxides present on the CeO₂ support. Figure 2 shows the La K-edge EXAFS spectra of LaMnO₃/CeO₂ catalysts along with those of the bulk LaMnO₃ and La single oxide (La₂O₃). The EXAFS spectrum of the bulk LaMnO₃ shows peaks at around 1.90 and 2.82 Ǻ, which were identified as the La–O and La–Mn contributions of the perovskite structure, respectively. The perovskite oxides have 12-coordinated La–O bonds with different bond lengths in their first coordination shells (2.50–3.01 Ǻ) and 8-coordinated La–Mn bonds in their second coordination shell (~3.37 Ǻ). The first coordination shell cannot be fitted by a single La–O bond, indicating that the shell was composed of several La–O bonds of different bond lengths. The peak in the second coordination shell, however, can be fitted by a single La–Mn bond. The spectrum of La–single oxide indicates a La–O contribution at 2.09 Ǻ and La–La contribution at 3.90 Ǻ, which were much different from the peak positions of the bulk LaMnO₃.

Fig. 2

La K-edge EXAFS spectra of LaMnO₃/CeO₂ catalysts and LaMnO₃

The formation of a perovskite oxide phase for the LaMnO₃/CeO₂ catalysts was confirmed by the EXAFS studies. The LaMnO₃/CeO₂ catalysts also showed peaks at the same positions as those of the bulk LaMnO₃ catalyst. The absence of peaks due to the La–O and La–La contributions in the spectra of the LaMnO₃/CeO₂ catalyst shows that there were no impurity phases in the supported catalysts. The curve fitting results for the EXAFS spectra of the LaMnO₃/CeO₂ catalysts are listed in Table 2. The peak for the La–Mn contribution at 282 Ǻ can be fitted well within the parameters of a single bond. The bond lengths for the La–Mn contribution were 3.340 Ǻ and were not influenced by the deposition method. The values were consistent with those for the bulk LaMnO₃ perovskite oxides. The CN for the La–Mn contribution in LaMnO₃/CeO₂-PD was close to that for bulk LaMnO₃, whereas the CN value was much lower for LaMnO₃/CeO₂-DI. Thus, the CN value for the perovskite oxide phase depends on the preparation method. Since the CN value is related to the crystalline sizes of LaMnO₃, this implies that the sizes of the crystalline LaMnO₃ of the LaMnO₃/CeO₂ catalyst were smaller than those of the bulk LaMnO₃ catalyst and LaMnO₃/CeO₂-PD.

Table 2 EXAFS curve-fitting results for the LaMnO₃ and LaMnO₃/CeO₂ catalysts

Figure 3 shows the XPS spectra of the LaMnO₃/CeO₂ catalysts and bulk LaMnO₃. In the XPS spectra ranges of Mn 2p (635–660 eV), we observed the spin–orbit splitting of Mn 2p_1/2 and Mn 2p_3/2 signals (Fig. 4a). For LaMnO₃, the peak position of the Mn 2p_3/2 signal was located at around 641.1 eV (Table 1) [28]. In the case of LaMnO₃/CeO₂ prepared by the PD method, we observed peaks for the Mn 2p signals at the same positions as those for the bulk LaMnO₃ catalyst. For LaMnO₃/CeO₂-DI, on the other hand, the Mn 2p_3/2 peak was located in the range of 642.1 ± 0.1 eV, which was higher by 1.0 eV than that of bulk LaMnO₃. The full width at half maximum (FWHM) for the Mn 2p_3/2 signal was comparable and larger than that for bulk LaMnO₃. This finding indicates that the oxidation state of Mn increased by the deposition of LaMnO₃ on CeO₂ by the DI method due to the strong interaction between LaMnO₃ and CeO₂.

Fig. 3

XPS Mn 2p (a) and Ce 3d (b) spectra of LaMnO₃/CeO₂ catalysts and LaMnO₃

Fig. 4

TEM images of LaMnO₃/CeO₂-DI. Bright-field image (a), DES mapping image of La (b), Mn (c), and Ce (d)

In the Ce 3d region, peaks due to the presence of Ce 3d_5/2 and Ce 3d_3/2 were observed at 880–920 eV (Fig. 3b). These peaks mainly consisted of 3d_5/2–3d_3/2 spin–orbit-split doublets characteristic of stoichiometric CeO₂ (Ce⁴⁺) [29]. The peak positions for the LaMnO₃/CeO₂ catalysts were the same as those for bulk LaMnO₃, indicating that the surface states of the exposed Ce species were unaffected by the deposition of LaMnO₃.

Table 1 lists the calculated Mn/Ce intensity ratios for the LaMnO₃/CeO₂ catalysts. The value was higher for the catalyst prepared by the DI method than that prepared by the PD method. Because the XPS peak intensity ratio of metal-support elements is related to the dispersion of the metal oxides on the support [30], we can conclude that the dispersion of the LaMnO₃ phase was higher for LaMnO₃/CeO₂-DI. On the other hand, the XPS intensity ratio for LaMnO₃/CeO₂-PD was comparable to that for the mechanically mixed LaMnO₃-CeO₂.

Figure 4 shows TEM images of LaMnO₃/CeO₂-DI. The bright-field image (Fig. 2a) indicated that the catalyst was composed of particles with sizes of 10–20 nm. The DES mapping images (Fig. 2b–d) revealed that La and Mn were distributed over the whole particles of CeO₂ support. These observations confirm that LaMnO₃ particles were dispersed on the CeO₂ for LaMnO₃/CeO₂-DI.

3.2 H₂-TPR Studies

TPR of the LaMnO₃/CeO₂ catalysts by H₂ was carried out to further investigate the structure and reducibility of the catalysts. Figure 5 shows the H₂-TPR profiles for the LaMnO₃/CeO₂ catalysts, bulk LaMnO₃, and CeO₂. Here, the H₂ consumption was normalized to 0.2 g and 0.8 g for LaMnO₃ and CeO₂, respectively, whereas the value was normalized to 1.0 g for LaMnO₃/CeO₂ catalysts. For the bulk LaMnO₃ catalyst, H₂ consumption was observed in temperature ranges of 200–400 °C and 700–900 °C. The peak in the lower temperature range was ascribed to the reduction of Mn⁴⁺ to Mn³⁺, and that in the higher temperature range to the reduction of Mn³⁺ to Mn²⁺ [10]. For the bulk CeO₂ support, reduction peaks were observed at 300–550 °C due to the reduction of surface Ce⁴⁺ and at higher temperatures due to the reduction of the bulk CeO₂ support [31]. In the case of LaMnO₃/CeO₂-PD, peaks appeared at around 136 °C. No peaks for the reduction of surface oxygen species on CeO₂ were observed, which had been observed at around 500 °C for bulk CeO₂. Quantitative analysis revealed that the amounts of H₂ consumed in the range of 50–550 °C in the LaMnO₃/CeO₂ catalyst were larger than the sum of the values consumed for the bulk catalysts (0.20 g-LaMnO₃ and 0.80 g-CeO₂). This reveals that the peak for the surface oxygen species on CeO₂ was shifted to a lower temperature by the LaMnO₃ deposition and indicates that the reactivity of the surface oxygen species on CeO₂ was improved. The peak maximum at around 800 °C was also slightly shifted to a lower temperature, indicating that the reactivity of the oxygen species on the catalysts and in the lattice were improved by the deposition of LaMnO₃ onto CeO₂.

Fig. 5

H₂-TPR profiles of LaMnO₃/CeO₂ catalysts and LaMnO₃. Sample weight 0.050 g, gas composition 5 %H₂–N₂ balance, gas flow rate 30 mL/min, heating rate 10 °C/min

In the case of LaMnO₃/CeO₂-DI, the peaks in the low temperature range (~400 °C) were much different from those of the bulk LaMnO₃ and LaMnO₃/CeO₂-PD: the H₂ consumption started in the same temperature range as that of LaMnO₃/CeO₂-PD, although the peak at 136 °C increased and the temperature of the reduction peak for Mn⁴⁺ to Mn³⁺ for LaMnO₃ was greatly lowered by the LaMnO₃ deposition on CeO₂. These findings indicate that, due to the strong interaction between LaMnO₃ and CeO₂, the reactivity of the oxygen species on the catalyst surface and in the lattice were improved by the deposition of LaMnO₃ onto CeO₂. The amount of H₂ consumed in the range of 50–550 °C was larger than that consumed with LaMnO₃/CeO₂-PD. These results also indicate that the amount of H₂ consumed is correlated with the dispersion of the perovskite oxides and that the interaction between LaMnO₃ and CeO₂ was controlled by changing the deposition method.

3.3 Catalytic Activity of CeO₂-Supported Perovskite Oxides

Figure 6 shows the catalytic activities of supported LaMnO₃ catalysts for NO oxidation in the temperature range of 200–500 °C, as compared with those for the bulk LaMnO₃ catalyst. For the purposes of comparison, the conversion curves for the equilibrium of NO oxidation (NO + 1/2O₂ ⇌ NO₂) were indicated. The LaMnO₃/CeO₂ catalyst exhibited steady-state activity and NO was quantitatively transformed to NO₂. The NO conversion increased with increased catalyst temperature up to 300 °C and decreased in the higher temperature range, following the equilibrium conversions. This indicates that the oxidation of NO to NO₂ is equilibrated with the backward reaction. Thus, the activity for NO oxidation was dominated by not only kinetics but also thermodynamics. The bulk LaMnO₃ catalyst exhibited lower NO conversion activity, with a maximum at 350 °C. The conversions of the CeO₂ catalyst were lower than those for the LaMnO₃ catalyst, with a maximum of 40 % conversions at 400 °C. A simple calculation shows that the activities of the LaMnO₃/CeO₂ catalyst cannot be explained in terms of the activities of bulk LaMnO₃ and CeO₂: the physical mixing of LaMnO₃ and CeO₂ would correspond to much lower conversions. The LaMnO₃/CeO₂-DI exhibited higher activity than the catalyst prepared by the PD method, even though the latter had a larger surface area. The rates for NO oxidation at 200 °C for LaMnO₃/CeO₂-DI and LaMnO₃/CeO₂-PD were 7.4 × 10⁻⁶ and 3.3 × 10⁻⁶ mol g⁻¹ min⁻¹, respectively. Thus, the high catalytic activity of LaMnO₃/CeO₂ prepared by the DI method was ascribed to the highly-dispersed LaMnO₃ species on CeO₂ and to the interaction between the catalytic materials. This result is probably due to the improved reactivity of the oxygen species of LaMnO₃ by deposition on CeO₂.

Fig. 6

Catalytic activities of LaMnO₃/CeO₂ catalysts and LaMnO₃ for oxidation of NO to NO₂. Dashed line refers to the dependence of equilibrium conversion reaction temperature. Reaction gas NO 500 ppm-O₂ 5 %-N₂ balance, catalyst weight 0.10 g, the gas flow 150 mL/min (W/F = 0.04 g s/mL)

Because the LaMnO₃/γ-Al₂O₃ catalyst has shown high catalytic activity in propane oxidation [25], the results described above prompted us to compare the catalytic properties of the LaMnO₃/CeO₂ catalyst with those of a LaMnO₃/γ-Al₂O₃ catalyst prepared by the incipient wetness method, in which LaMnO₃ is deposited inside the alumina pores. The LaMnO₃/CeO₂ catalyst showed higher activity than the LaMnO₃/γ-Al₂O₃ catalyst (Fig. 6), even though the surface area of the LaMnO₃/CeO₂ catalyst was much smaller than that of the LaMnO₃/γ-Al₂O₃ catalyst (Table 1).

Lastly, we investigated the thermal stability of the LaMnO₃/CeO₂ catalyst for NO oxidation. The catalysts were heated at 850–1050 °C for 5 h in air and were again used for NO oxidation. As shown in Fig. 7, the X-ray diffraction (XRD) patterns of the LaMnO₃/CeO₂ catalyst exhibited the same CeO₂ structure as calcined catalysts when the catalyst was calcined at 850 °C. As the calcination temperature increased, the peak intensities increased and the FWHM decreased, which indicates the sintering of the supports. Correspondingly, the catalyst surface area significantly decreased: the surface areas of the LaMnO₃/CeO₂ catalysts after calcination at 850, 950, 1050 °C were 24, 11, 2 m²/g, respectively. On the other hand, while the peaks for the perovskite oxides became more prominent, no other impurity phases were observed. It is worth noting that the calcination temperature in the range of 650–1050 °C was suitable for preparing bulk LaMnO₃ perovskite powder, and the supported LaMnO₃ phase was formed in the same temperature range.

Fig. 7

XRD patterns of LaMnO₃/CeO₂-DI calcined at different temperatures

The NO oxidation activity in the thermally treated LaMnO₃/CeO₂ catalyst decreased with increased catalyst calcination temperature. The maximum temperature for NO oxidation shifted to a higher temperature and the maximum conversion decreased: the conversions slightly decreased after the calcination temperatures reached 850–950 °C, and significantly decreased after the calcination temperature reached 1050 °C (Fig. 8). We also investigated the themal stability of LaMnO₃/γ-Al₂O₃ catalysts because they showed high stability for propane oxidation even when heated at high temperatures (650–1050 °C). Although the NO oxidation activity of the LaMnO₃/γ-Al₂O₃ catalyst decreased after themal treatment at temperatures between 850–1050 °C, the decrease in the NO oxidation activity of these catalyst was smaller than in the LaMnO₃/CeO₂ catalysts. This is probably due to the improved stability of the perovskite phases when they were deposited inside the γ-Al₂O₃ pores. A comparison of the activity of the LaMnO₃/CeO₂ catalyst with that of LaMnO₃/γ-Al₂O₃, showed that the LaMnO₃/CeO₂ catalyst exhibited higher activity after calcination at 850 °C, whereas this activity dropped significantly after calcination at 1050 °C. This decrease in the catalyst activity was probably due to the decreased surface area and catalyst sintering. Thus, the LaMnO₃/CeO₂ catalyst is effective in promoting NO oxidation when calcined in the temperature range of 650–850 °C.

Fig. 8

Effect of heating treatment of LaMnO₃/CeO₂ and LaMnO₃/γ-Al₂O₃, catalysts on their NO oxidation activities. The reaction condition was the same as that in Fig. 6

4 Conclusion

In this study, we prepared LaMnO₃/CeO₂ catalysts with different perovskite dispersions using four methods. The formation of LaMnO₃ perovskite phases was confirmed by EXAFS studies. The dispersion of LaMnO₃ perovskite oxide was evaluated by XPS. The LaMnO₃/CeO₂ catalyst prepared by the incipient wetness method exhibited higher NO oxidation activity although this catalyst has the smallest surface area. Based on the H₂-TPR study results, we propose that the higher level of activity may be ascribed to the strong interaction between the perovskite phase and the CeO₂ support. The supported catalyst was calcined in the temperature range of 650–1050 °C and then used for NO oxidation because thermal stability of the catalyst is important from a practical application perspective. The LaMnO₃/CeO₂ catalyst dry-impregnation method was more stable than LaMnO₃/γ-Al₂O₃ catalyst against heat treatment at the temperature lower than 950 °C.