Effect of the Mn oxidation state and lattice oxygen in Mn-based TiO₂ catalysts on the low-temperature selective catalytic reduction of NO by NH₃

Abstract

TiO₂-supported manganese oxide catalysts formed using different calcination temperatures were prepared by using the wet-impregnation method and were investigated for their activity in the low-temperature selective catalytic reduction (SCR) of NO by NH₃ with respect to the Mn valence and lattice oxygen behavior. The surface and bulk properties of these catalysts were examined using Brunauer-Emmett-Teller (BET) surface area, X-ray diffraction (XRD), temperature-programmed reduction (TPR), and temperature-programmed desorption (TPD). Catalysts prepared using lower calcination temperatures, which contained Mn⁴⁺, displayed high SCR activity at low temperatures and possessed several acid sites and active oxygen. The TPD analysis determined that the Brönsted and Lewis acid sites in the Mn/TiO₂ catalysts were important for the low-temperature SCR at 80∼160 and 200∼350 °C, respectively. In addition, the available lattice oxygen was important for attaining high NO to NO₂ oxidation at low temperatures.

Implications:

Recently, various Mn catalysts have been evaluated as SCR catalysts. However, there have been no studies on the relationship of adsorption and desorption properties and behavior of lattice oxygen according to the valence state for manganese oxides (MnO_x). Therefore, in this study, the catalysts were prepared by the wet-impregnation method at different calcination temperatures in order to show the difference of manganese oxidation state. These catalysts were then characterized using various physicochemical techniques, including BET, XRD, TPR, and TPD, to understand the structure, oxidation state, redox properties, and adsorption and desorption properties of the Mn/TiO₂ catalysts.

Introduction

Selective catalytic reduction (SCR) using NH₃ is the most widely used method of removing nitrogen oxides (NO_x) emitted from stationary sources (Bosch and Janssen, 1988; Parvulescu et al., 1998). V/TiO₂ or V/W/TiO₂ catalysts have been commercialized for SCR and operate under moderate temperatures ranging from about 300 to 400 °C (Alemany et al., 1995; Centi and Perathoner, 1995). SCR reactors at power plants and gas turbines can be located in the later portions of the dust collection and desulfurization devices to avoid deactivation from dust or SO₂. At this point in the process, the temperature is less than 200 °C, which is where the conventional V/W/TiO₂ catalysts used in the NH₃-SCR process cannot be used (Kuo et al., 1997; Li et al., 2008). Accordingly, a denitrification catalyst is required for successful SCR operation at temperatures under 200 °C. In this temperature range, MnO_x-based catalysts are suitable for the NH₃-SCR system. Supported manganese oxides such as MnO_x/Al₂O₃ (Stoilova et al., 1998) MnO_x/TiO₂ (Li et al., 2007), MnO_x/CeO₂ (Qi and Yang, 2003), and MnO_x/active carbon (Marbán et al., 2004) possess excellent activity in low-temperature SCR. These catalysts have been shown to affect the SCR activity of Mn-based catalysts through changes to the Mn oxidation state. This result has been previously demonstrated by many researchers. We have also previous reported that natural manganese oxides readily undergo the redox cycle and the oxidation state of manganese ions can be easily changed (Kim et al., 2011). Kapteijn et al. (1994) investigated the reduction of NO by NH₃ over pure manganese oxides and found that the NO conversions decreased in the order of MnO₂ > Mn₅O₈ > Mn₂O₃ > Mn₃O₄. Tang et al. (2006) prepared a Mn/CeO₂ catalyst using a modified co-precipitation method and tested for its HCHO oxidation activity. They reported that the MnO_x-CeO₂ catalyst that was calcined at 500 °C possessed more Mn⁴⁺ and richer lattice oxygen species, which resulted in significantly higher catalytic activity. Peña et al. (2004) reported that Mn/TiO₂ catalysts prepared at lower calcination temperatures provided the best catalytic performance, and the lower reduction temperature produced the MnO₂ and Mn₂O₃ phases. However, the effect of the manganese oxidation state on the acid site and lattice oxygen has not been clearly demonstrated. Roy et al. (2008) reported that NH₃ can be adsorbed onto the surface of the catalysts as NH₄ ⁺ and molecular NH₃ at the Brönsted and Lewis acid sites, respectively. The adsorbed NH₃ and NH₄ ⁺ can form an NH₂NO adduct with adsorbed NO, which dissociates into N₂ and H₂O. Peña et al. (2004) reported that the Lewis acid sites of the catalyst appear to play an important role in activating NH₃ in low-temperature SCR involving TiO₂, such as Mn/TiO₂ and Cu/TiO₂. The purpose of the present study is to better understand the relationship between the acid sites, valence of manganese, and lattice oxygen. To achieve this goal, the catalysts were prepared by the wet-impregnation method at different calcination temperatures to produce manganese catalysts with different oxidation states. These catalysts were characterized using various physicochemical techniques, including Brunauer-Emmett-Teller (BET) surface area, X-ray diffraction (XRD), temperature-programmed reduction (TPR), and temperature-programmed desorption (TPD), to understand the structure, oxidation state, as well as redox, adsorption, and desorption properties of the Mn/TiO₂ catalysts.

Experimental

Catalyst preparation

Mn/TiO₂ catalysts were prepared using wet impregnation at different calcination temperatures with an MC-50 support. MC-50 is a commercial anatase TiO₂ provided by Ishihara Sangyo Kaisha (Ltd., Japan). The required amount (20 wt% Mn w/w) of manganese(II) nitrate hydrate (Mn(NO₃)₂∙6H₂O; Aldrich Chemical; Co. Inc., Korea) was dissolved in distilled water at 80 °C. After impregnation, the moisture was evaporated at 70 °C using a rotary evaporator and dried at 110 °C overnight. The catalysts were calcined for 4 hr at temperatures ranging from 300 to 700 °C. The obtained samples were ground and sieved through 40–50 mesh. The following naming convention was used to indicate the prepared catalysts: Mn/TiO₂ (300) denotes 20 wt% Mn/TiO₂ (MC-50) calcined at 300 °C.

Characterization

The surface areas of the TiO₂ and manganese-loaded TiO₂ catalyst powders were measured by physical nitrogen adsorption at −196 °C using an ASAP 2010 (Micrometrics Inc., USA). Approximately 0.1∼0.3 g of the catalysts were degassed at 300 °C for 2 hr.

X-ray diffraction (XRD) measurements were conducted using Cu Kα (λ = 1.5056 Å) radiation. The catalysts were scanned over a 2θ range of 20°∼90° with a step size of 0.1° at 1.0-sec intervals using X'Pert PRO MRD (PANalytical Spectrics Korea Ltd., Korea).

The temperature-programmed reduction (TPR) of H₂ in 10% H₂/Ar was measured using 0.3 g of catalyst with a total flow rate of 50 cc/min. The catalyst was pretreated in a 5% O₂/Ar flow at 300 °C for 0.5 hr and cooled to 50 °C before measuring the H₂-TPR. The catalyst was placed in dilute hydrogen, and the hydrogen consumption was monitored using an Autochem 2920 (Micrometrics) by increasing the temperature to 800 °C at a rate of 10 °C/min. NH₃ temperature-programmed desorption (TPD) measurements were performed using 30 mg of the sample with a total flow of 50 cc/min. Before the TPD measurements, the catalyst was pretreated under a 5% O₂/Ar flow at 300 °C for 0.5 hr and cooled to 50 °C. The samples were treated with 1% NH₃/Ar for 0.5 hr. Adsorbed NH₃ was removed using Ar flow for at least 1.5 hr before starting the TPD experiments. The NO-TPD (1% NO/Ar) experimental conditions were identical to the NH₃-TPD experiment. During the TPD experiments, NH₃ (m/z 15), NO (m/z 30), N₂O (m/z 44), N₂ (m/z 28), NO₂ (m/z 46) were continuously monitored using a quadrupole mass spectrometer (QMS 422; Pfeiffer-vaccum., Germary) while increasing the temperature to 600 °C at a rate of 10 °C/min.

Catalytic activity measurement

The SCR apparatus consisted of a continuous-flow-type fixed-bed reaction system made of a quartz tube (inner diameter: 8 mm and height: 650 mm) and a catalytic bed that was fixed using quartz wool. The catalysts were made using a hydraulic press to pellet with 5,000 pounds of force and a 40–50-mesh sieve. It was then charged in a quartz tube. The temperature of the reactor was controlled using a proportional-integral-derivative (PID) controller and a K-type thermocouple fixed on the bottom of the bed. In order to measure the gas inlet temperature, another K-type thermocouple was installed on the top of the catalytic bed. The feeding gases consisted of 200 ppm NO_x, 220 ppm NH₃, 8% O₂, and 8% moisture. Two hundred and eighty milligrams of sample was used in each run. The total flow through the reactor was 500 cc/min and a gas hourly space velocity (GHSV) of 60,000 hr⁻¹ was obtained. The premixed gases were supplied by a mass flow controller (MFC; MKS Co. Ltd., Korea). Moisture was fed into the reactor by injecting N₂-containing water vapor through the bubbler, and this level of moisture was maintained constant by circulating water at a fixed temperature of 45 °C using a circulator outside the jacket-type bubbler. The entire gas supply pipe was made of stainless steel and was wrapped with a heating band set at 180 °C so that the moisture would not condense. The concentrations of the reactants and products were measured as follows: inlet and outlet NO and N₂O concentrations were analyzed using a NO_x analyzer (Fuji Electric Co., Ltd., Japan) and N₂O analyzer (Ultramat 6; Siemens Co., Germany). The concentration of NO₂ and NH₃ was measured using two different detector tubes: 9L (GASTEC Co., Japan) and 3M, 3La, 3L (GASTEC Co., Japan), respectively.

Results and Discussions

Characterization of catalysts

The surface area, pore volume, and pore size of the Mn/TiO₂ catalysts that were prepared using different calcination temperatures are shown in Figure 1. The surface area of the TiO₂ support was 66.19 m²/g (not shown). As the calcination temperature increased from 300 to 700 °C, the surface area and pore volume decreased from 60 to 8 m²/g and 0.35∼0.1 cm³/g, respectively. In contrast, the pore diameter increased from 21 to 40 nm, which was expected. The XRD patterns for these catalysts are shown in Figure 2. MnO₂ peaks were not observed at 28.9°, 37.7°, and 57.1° in any catalyst. However, when the calcination temperature was above 600 °C, Mn₂O₃ peaks were observed at 23.0°, 32.9°, and 45.1°. In addition, the anatase TiO₂ phase was partially transformed into the rutile TiO₂ structure for the Mn/TiO₂ catalyst calcined at 700 °C. Qi et al. (2004) reported that Mn₂O₃ and MnO₂ were detected by XRD when pure manganese was calcined at 700 °C. In this study, amorphous MnO₂ was formed because of the high surface energy for manganese in the TiO₂ support. Sharp peaks were observed for the crystallite Mn₂O₃ structure when the catalyst was calcined at a temperature above 600 °C. For catalysts calcined at temperatures between 300 and 500 °C, no manganese structures were detected by XRD analysis, which may suggest that either the manganese was in a highly dispersed state or that the manganese ions had been inserted into the TiO₂ lattice (Pechi et al., 2002). However, it is thought that the catalysts calcined at temperatures over 600 °C plugged the micropores of the support materials and either sintered or agglomerated as crystallite Mn₂O₃ because of the thermal effect. These results are in agreement with previous studies (Peña et al., 2004; Smirniotis et al., 2006).

Figure 1. Effect of calcination temperature on surface area, pore volume, and pore diameter of the Mn/TiO₂ catalysts.

Figure 2. X-ray diffraction patterns of the Mn/TiO₂ catalysts with different calcination temperatures (A = anatase TiO₂; R = rutile TiO₂).

TPR analysis

The XPS and H₂-TPR analysis were mainly used to evaluate the oxidation state of manganese oxide. In recent literature, the SCR performance of all the Mn/TiO₂ catalysts is accurately associated with the surface Mn⁴⁺ concentrations by XPS analysis (Thirupathi and Smirniotis, 2012). Also, these authors reported that the MnO₂ phase is extremely dominant over the Mn₂O₃ phase (Mn⁴⁺/Mn³⁺ = 22.31, 96%) in the Mn-Ni(0.4)/TiO₂ anatase catalyst and H₂-TPR results are in good accordance with XPS results (Thirupathi and Smirniotis, 2011). In this paper, H₂-TPR experiments were performed to investigate the oxidation state of manganese (Figure 3). Ettireddy et al. (2007) reported that peak reductions for the 11.1 wt% Mn/TiO₂ catalyst occurred at 215, 331, 421, and 468 °C. The first peak change at 215 °C was from the reduction of Ti⁴⁺ to Ti³⁺. In the second reduction step, which occurred at 331 °C, MnO₂ was reduced to Mn₂O₃, which was further reduced to Mn₃O₄ and finally to MnO at 421 and 468 °C, respectively. In this study, the calcined TiO₂ contained only one reduction peak near 600°C, which was very low relative to the TiO₂-supported manganese oxide catalysts. The reduction of Ti⁴⁺ to Ti³⁺ did not occur for any of the catalysts at 215 °C. For the Mn/TiO₂ (300) catalyst, the first reduction of MnO₂ to Mn₂O₃ occurred at 303 °C, and the second reduction step of Mn₂O₃ to Mn₃O₄ occurred at 477 °C. The last reduction step, which converted Mn₃O₄ to MnO, occurred at 546 °C. However, unlike the other catalysts, the Mn/TiO₂ (300) catalyst possessed overlapping reduction peaks in the area of the Mn₂O₃ to Mn₃O₄ transition at 615 °C. This extra peak may be due to the slight reduction of Ti⁴⁺ to Ti³⁺ in the bulk TiO₂ after the surface manganese species had been reduced. Peña et al. (2004) reported that higher reduction temperatures were required for catalysts synthesized with lower Mn loading or calcined at higher temperatures. It was also reported that a catalyst calcined at higher temperatures (>400 °C) could not be easily reduced because of the stronger metal-support interactions (Peña et al., 2004). Manganese catalysts calcined at 400 °C possessed a similar reduction peak as catalysts calcined at 300 °C. However, the overlapping peak at approximately 615 °C was not observed. Increasing the calcination temperature from 400 to 600 °C shifted the reduction peak to lower temperatures, and the MnO₂ to Mn₂O₃ reduction peak was not observed in catalysts calcined above 600 °C. These results indicate that the primary Mn species in the Mn/TiO₂ (600) and Mn/TiO₂ (700) catalysts consisted of Mn³⁺. The Mn/TiO₂ (700) catalyst only displayed a single, low-intensity peak for the reduction of Mn₂O₃ to Mn₃O₄, which occurred at 477 °C. This peak was shifted to a lower temperature compared with the other catalysts. The XRD results confirmed that the transition from the anatase phase to the rutile phase occurred in the Mn/TiO₂ (700) catalyst because of the thermal effect. The interaction between the MnO_x and anatase-phase TiO₂ was stronger than in the rutile phase (Ettireddy et al., 2007), and the reduction transition occurred at a different temperature can be explained by the presence of different TiO₂ phases (Ettireddy et al., 2007). It indicated that the interaction of manganese with rutile TiO₂ is weaker than the anatase TiO₂. Therefore, the reduction of Mn₂O₃ to Mn₃O₄ in the Mn/TiO₂ (700) catalyst occurred at a lower temperature compared with the other catalysts.

Figure 3. H₂-TPR profiles of the Mn/TiO₂ catalysts with different calcination temperatures.

In addition, as shown by the BET result, severe induration was caused by the plugging of small pores, which left only large pores and reduced diffusion. Thus, the Mn/TiO₂ (700) catalyst has a greater difficulty removing oxygen, and the last reduction step of Mn₃O₄ to MnO did not occur. The TPR and XRD results indicated that most of the manganese was stabilized in the Mn⁴⁺ oxidation state for the Mn/TiO₂ (300–500) catalysts. In contrast, most of the manganese was stabilized in the Mn³⁺ oxidation state for the Mn/TiO₂ (600–700) catalysts.

TPD analysis

NH₃-TPD experiments using TCD were performed to investigate the adsorption properties of NH₃ (Figure 4). Three desorption peaks were observed between about 80 and 600 °C for the Mn/TiO₂ (300–500) catalysts. In contrast, broad desorption peaks were observed between about 80 and 500 °C for the Mn/TiO₂ (600–700) catalysts. Several studies on the Brönsted and Lewis acid sites in Mn/TiO₂ catalysts have been reported, and NH₃ adsorbed onto the Brönsted acid sites desorbed at a lower temperature than the Lewis acid–coordinated NH₃. Many researchers have reported that Lewis acidity is important in low-temperature SCR; however, the reported NH₃ desorption temperatures are quite different when using a TCD detector. For example, Chmielarz et al. (2004) reported that NH₃ was desorbed from the strong and broad Lewis acid sites at approximately 650 °C, and Jin et al. (2010) reported that NH₃ desorbed from a Mn-Ce/TiO₂ catalyst at 570∼680 °C and was most strongly linked to the Lewis acid sites. Wu et al. (2008) reported that NH₃ adsorbed at the weak and medium acid sites desorbed at 100∼350 °C, whereas the NH₃ adsorbed at strong Lewis acid sites desorbed between 564 and 665 °C. The Mn/TiO₂ catalyst is one of the best catalysts for oxidizing NH₃ (Wu et al., 2008). Therefore, it is possible that the strongly chemisorbed NH₃ was oxidized by lattice oxygen during the NH₃-TPD experiment.

Figure 4. NH₃-TPD profiles of the Mn/TiO₂ catalysts with different calcination temperatures using TCD.

NH₃-TPD experiments that use TCD are capable of measuring all materials in addition to NH₃. Therefore, NH₃-TPD experiments using a mass spectrometer were performed to investigate NH₃ and other by-products (Figure 5). NH₃ was recorded using the molecular weights of 15, 16, 17, and 18 and was measured simultaneously with oxygen (m/z 16) and moisture (m/z 17 and 18). With the exception of catalyst calcined at 700 °C, all of the catalysts contained two broad NH₃ desorption peaks at 80∼160 and 200∼350 °C. These two peaks were attributed to the Brönsted and Lewis acid sites. As shown in Figure 5B, NO, NO₂, and N₂O were generated through the oxidation of NH₃, which was attributed to the desorption of NH₃ chemisorbed to strong Lewis acid sites at approximately 200∼350 °C. The lower the calcination temperature, the greater the amount of oxidized material, which could only be from the lattice oxygen present in the catalyst because gaseous oxygen was not provided. The relative quantity of materials oxidized by the lattice oxygen in the catalyst was very low relative to NH₃, and the lower calcination temperatures produced more lattice oxygen. In addition, the peak formed at 500 °C in the mass analysis corresponded to oxygen (m/z 16) and not NH₃ (m/z 15). Because of the TPR test, the reduction of MnO₂ to Mn₂O₃ in the Mn/TiO₂ (300–500) catalysts was shown to occur at a temperature of 303 °C. Accordingly, the activation of oxygen plays a role in partially oxidizing the adsorbed NH₃ through the reduction of MnO₂ to Mn₂O₃, which occurred at approximately 300 °C. In addition, the structure of Mn–O–Ti is unstable and provides further lattice oxygen molecules for the oxidation of NH₃. As this structure is unstable, Mn₂O₃ might be reduced to Mn₃O₄ when the temperature is over 500 °C because of the thermal effect, and the oxygen would not participate in reactions within the developed catalysts. Mn/TiO₂ (600–700) catalysts were already calcined at temperatures over 600 °C, and their manganese oxidation state on the catalyst surface does not form an unstable structure with deficient lattice oxygen and mostly exists as Mn³⁺. Therefore, the generation of oxygen was not observed for the Mn/TiO₂ (600–700) catalysts at temperatures over 500 °C during the NH₃-TPD test. For such catalysts, the reduction of MnO₂ to Mn₂O₃ at approximately 300 °C did not occur. As the amount of oxygen available was insignificant because of the reduction of MnO₂ to Mn₂O₃, NH₃ oxidation using lattice oxygen was incomplete. Because the NH₃-TPD test used Mn/TiO₂ catalysts prepared at different calcination temperatures, the chemisorptions at the Lewis acid sites for temperatures over 200 °C were shown to be more important than desorption from the Brönsted acid sites at approximately 100 °C in the low-temperature SCR. In addition, for catalysts calcined at lower temperatures, the acidity of the acid site might change to allow for the use of the lattice oxygen, which had a higher catalytic activity due to the effective activation of molecular oxygen through the oxygen transfer mechanism.

Figure 5. NH₃-TPD profiles of the Mn/TiO₂ catalysts with different calcination temperatures using mass spectrum. (A) Investigation of NH₃, O₂ and H₂O; (B) investigation of N₂, NO, N₂O, and NO₂ (color figure available online).

NO-TPD experiments using a mass spectrometer were also performed to investigate the adsorption properties of nitrogen oxide. All of the catalysts contained two desorption peaks (Figure 6). The first peak corresponded to either a weakly or physically adsorbed NO and appeared at 100 °C, whereas a broader second peak, which corresponded to desorption of the more strongly bound nitrate species, occurred between 200 and 300 °C. No desorbed NO species were observed above 300 °C. A similar NO desorption trend was observed for all of the catalysts. The relative quantity of NO desorbed from the Mn/TiO₂ (600–700) catalysts was relatively small. As shown in Figure 6B, NO₂ began to desorb at approximately 150 °C, with the maximum desorption occurring between 250 and 300 °C. The results of the NO-TPD test were similar to the NH₃-TPD test. Most of the NO₂ was oxidized from NO through chemically adsorbed oxygen, and the higher the oxidation state of the Mn catalyst, the more NO₂ was produced. In addition, unlike the catalysts calcined at 300 and 400 °C, which reached a maximum peak at 270 °C, the peak maximum for catalysts calcined at 500 °C was 278 °C, and the maximum peak for catalysts calcined at 600 °C was 288 °C. As the calcination temperature decreased, the Mn⁴⁺ content of the catalyst increased, which could be easily identified using the lattice oxygen. The low-temperature SCR continued once NH₃ and NO were adsorbed. Because of the TPD test, both adsorbed compounds (NH₃ and NO) were shown to produce lattice oxygen. The oxidation of NH₃, a reducing agent in the SCR, negatively affected SCR activity. However, the materials formed from this oxidation were primarily N₂ and NO₂, whereas during the fast SCR process, only some of the NO was oxidized to NO₂, the reducing agent was oxidized, and the oxidation process did not have a profound effect on SCR activity. Therefore, the capability to oxidize NO to NO₂ plays a greater role than NH₃ in the SCR.

Figure 6. NO-TPD spectra of the Mn/TiO₂ catalysts with different calcination temperatures using mass spectrum. (A) Investigation of NO; (B) investigation of NO₂.

NO oxidation activity

In the previous TPD test, the NH₃ and NO adsorption characteristics of Mn/TiO₂ catalysts prepared at different calcination temperatures were evaluated. A similar trend of desorption was observed for all of the catalysts, except the catalyst calcined at 700 °C, where a higher calcination temperature resulted in a reduced production of the oxidizing material. When both NO and NO₂ were included into the feed, a fast SCR may have occurred when compared with the standard SCR of NO (Ciardelli et al., 2007; Koebel et al., 2002a, 2002b; Tronconi et al., 2007). The oxidation activity tests for the conversion of NO to NO₂ by O₂ on the Mn/TiO₂ catalysts prepared at different calcination temperatures were performed (Figure 7). The lower the calcination temperature, the higher the NO oxidation activity, and a conversion rate of 76% was obtained at 250 °C. The adsorptive capabilities of NO in the region of low temperatures were similar to each other; the cause for the difference in NO oxidation activity with the same amount of gaseous oxygen was due to the difference in the amount of lattice oxygen that participated in the reaction inside the catalyst. The Mn/TiO₂ catalysts can be reduced from MnO₂ to Mn₂O₃, and the released oxygen species from MnO₂ participated in the oxidation of NO to NO₂. The reduced Mn₂O₃ can then be reoxidized by gaseous oxygen. That is to say, gaseous oxygen cannot be expected to participate in the reaction at extremely low temperatures, but it can participate in reoxidation.

Figure 7. Oxidation activity of NO to NO₂ by O₂ on the Mn/TiO₂ catalysts with different calcination temperatures. (Reaction condition: 200 ppm NO, 8% O₂, 6% H₂O, GHSV = 60,000 hr⁻¹).

SCR activity

Figure 8 shows the NO_x activity tests for the Mn/TiO₂ catalysts with respect to their calcination temperatures. Conversion of NO_x decreased with increasing calcination temperature for all of the catalysts. At low temperatures, the Mn/TiO₂ catalyst calcined at 300 °C showed the highest NO_x removal efficiency, which was approximately 90% of the conversion rate at 180 °C. However, the crystalline Mn₂O₃ catalyst did not convert NO_x at any temperature.

Figure 8. NO_x conversion of the Mn/TiO₂ catalysts with different calcination temperatures. (Reaction condition: 200 ppm NO_x, 8% O₂, 6% H₂O, NH₃/NO_x = 1, GHSV = 60,000 hr⁻¹).

Conclusions

The oxidation state and lattice oxygen behavior of the Mn/TiO₂ catalysts strongly affected their SCR activity at low temperatures. The structure and oxidation state analyses by XRD and TPR confirmed that the manganese oxidation state (Mn⁴⁺) was higher for the Mn/TiO₂ (300–500) catalysts than for the Mn/TiO₂ (600–700) catalysts. TPD analysis revealed the presence of Brönsted and Lewis acid sites at 80∼160 and 200∼350 °C, respectively. The Lewis acid sites, which chemisorbed at temperatures above 200 °C, were important to the low-temperature SCR. However, regardless of the manganese oxidation state on the catalyst surface, the adsorption and adsorptive strength of NH₃ and NO did not change. Catalysts calcined at low temperatures existed as amorphous MnO₂ and contained active lattice oxygen in more abundance than catalysts containing Mn₂O_3. As indicated by the NO oxidation test, the oxygen released from the MnO₂ structure can more readily oxidize NO to NO₂ than the Mn₂O₃ structure, which resulted in a higher NO_x conversion rate because of this excellent redox characteristic. The manganese oxidation state and availability of lattice oxygen were shown to play important roles in low-temperature SCR activity.

Acknowledgments

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2010-0023963).