Nitrogen oxides (NO, NO2, and N2O) from plant emissions and vehicle exhausts cause various environmental problems such as photochemical smog, acid rain, ozone depletion, and greenhouse effects. Among the methods exploited for the elimination of NOx, selective catalytic reduction (SCR) of NOx with NH3 has been confirmed as the most effective approach. Two types of commercial catalysts, V2O5/TiO2 catalysts promoted by WO3 or MoO3 [1, 2, 3, 4] and transition-metal exchanged zeolites [3, 4, 5, 6], have been applied to this process. Non-zeolitic oxides [4, 7] have also been developed for the removal of NOx at low temperatures (100-300 °C). In recent years, CeO2 based catalysts have attracted much attention owing to their oxygen storage and redox properties via the redox shift between Ce4+ and Ce3+ [8].
The low-temperature activity of MnOx-CeO2 catalyst for SCR of NOx with NH3 was first reported by Qi et al. [9, 10, 11], who initiated extensive studies in the last decade. In the catalyst, CeO2 acts mainly as a support for dispersion of and interaction with MnOx, leading to more facile active oxygen species. The influences of various dopants, additives, and supports, including the oxides of Nb [12, 13], W [12, 14, 15], Ti [16], Zr [12, 17], Sn [18], Ca [19], and Fe [12, 20] and carbon nanotubes [21], on the deNOx activity of MnOx-CeO2 catalysts have been widely studied. In these reports, the enhanced low-temperature SCR activities of such modified MnOx-CeO2 catalysts are generally ascribed to improved redox properties and increased acidity. However, controversial effects of a third component have often been obtained in different studies, depending on the Mn/Ce ratio in the catalyst and the preparation method adopted. For example, Casapu et al. [12] observed a drastic decrease in SCR activity when Fe2O3 was added to MnOx-CeO2 by coprecipitation. In contrast, Shen et al. [20] reported that Fe doping by sol-gel enhanced the catalytic activity and H2O and SO2 resistance of MnOx-CeO2/TiO2, which is ascribed to increased specific surface area and NH3 and NOx adsorption capacity, as well as improved dispersion and oxidation states of Mn and Ce on the catalyst surface. Additionally, the still modest activity of MnOx-CeO2 catalysts at relatively high temperature and their particularly low selectivity require improvement.
High N2 selectivity for SCR over MnOx-CeO2 catalysts can be obtained by adding an acidic component, but this simultaneously results in a decrease in NH3-SCR activity at low temperatures. A comprehensive balance between redox sites and acid sites is critical to broadening the temperature window of these catalysts. Casapu et al. [13] achieved a good distribution of oxidizing and acidic sites in the MnNbCe catalyst structure and, through formation of MnNb2O6, diminished the oxidation ability of the Mnx+ species. As a consequence, the unselective oxidation of NH3 was avoided at high temperatures. Nevertheless, an excess of Mn relative to the Nb content is needed for low-temperature SCR activity. Thus, it is important to further clarify the relationship between these two important properties of MnOx-CeO2 catalysts by adding different acidic components.
In this study, MMnCe (M = Nb, W, and Mo) oxide catalysts were synthesized using a sol-gel method. The temperature window of the catalysts was expected to be controllable via the redox properties of the MnOx-CeO2 mixed oxides and the surface acidity of MOx. The interactions between MOx and MnOx- CeO2 and the corresponding effects on the NH3-SCR mechanisms were elucidated.
MnCe (with a molar ratio of Mn:Ce = 4:6) and MMnCe (M = Nb, W, and Mo, with a molar ratio of M:Mn:Ce = 1:4:5) catalysts were prepared by a citric acid-aided sol-gel method. Mn(NO3)2 (Beihua, Beijing, China), Ce(NO3)3·6H2O (Yili, Beijing, China), C12H7NbO24 (Alladin, Shanghai, China), (NH4)10·W12O36·5H2O (Beihua), and (NH4)6Mo7O24·4H2O (Beihua) were used as the precursors. Citric acid was added as the complexing agent at twice the amount of the metal ions. Polyglycol was then added at 10 wt% of the citric acid. The solution was magnetically stirred and heated at 70 °C until a porous gel was formed. The obtained gel was dried at 110 °C overnight followed by decomposition at 300 °C for 1 h and calcination at 600 °C for 3 h in a muffle furnace.
Measurement of the activity of the catalysts for NOx reduction with NH3 was carried out in a fixed-bed reactor made of a quartz glass tube. The reaction gas mixture consisted of 500 ppm NO, 500 ppm NH3, 5% O2, and N2 in balance. The measurement was performed from 100 to 350 °C with a temperature interval of 50 °C. Two hundred milligrams of the catalyst was sieved to 40−80 mesh and then diluted to 1 mL with SiO2 pellets using a gas hourly space velocity (GHSV) of 30000 h−1. The concentrations of nitrogen oxides (NO, NO2, and N2O) and NH3 were measured with a Nicolet 380 FTIR spectrometer (Thermo Fisher, WI, USA) equipped with a 2 m path-length sample cell. The NOx conversion and N2 selectivity of the catalyst were calculated as follows:
NOx conversion (%) = (NOin − NOout − NO2 out)/NOin × 100
Selectivity (%) = (1 − (NO2 out + 2N2Oout)/(NOin − NOout + NH3 in − NH3 out)) × 100
Similar reaction conditions were adopted for the NH3 (or NO) oxidation experiments, for which NO (or NH3) was not introduced in the gas mixture.
X-ray diffraction (XRD) patterns of the catalysts were determined using a D/mas-RB diffractometer (Rigaku, Tokyo, Japan) employing Cu Kα radiation (λ = 0.15418 nm). The X-ray tube was operated at 40 kV and 120 mA. The XRD patterns were recorded at 0.02° intervals in the range 20° < 2θ < 80°. Identification of the phases was made with the help of JCPDS (Joint Committee on Powder Diffraction Standards) cards. The mean crystallite size of CeO2 in the samples was calculated using the Williamson-Hall equation.
The specific surface areas of the samples were measured from their N2 adsorption isotherms at −196 °C by the four-point Brunauer-Emmett-Teller (BET) method using an automatic surface analyzer (F-sorb 3400, Gold APP Instrument, Beijing, China). The samples were degassed in flowing N2 at 200 °C for 2 h.
H2 temperature-programmed reduction (H2-TPR) was performed on an Auto Chem II (Micromeritics Instrument Corp., GA, USA). Prior to the analysis, 50 mg of sample was treated in a 50 mL/min flow of He at 500 °C for 30 min. The reactor temperature was raised to 900 °C at a heating rate of 10 °C/min in 10% H2/He (50 mL/min).
Temperature-programmed desorption of NOx (NOx-TPD) was carried out in a fixed-bed reactor with the emission gases determined using a Thermo Nicolet 380 FTIR spectrometer. Powder catalyst (0.2 g) was diluted to 1 mL with silica and preheated in 5% O2/N2 at 500 °C for 30 min. After cooling to RT, the sample was exposed to a flow of 1000 ppm NO in 5% O2/N2 for 30 min and then flushed with N2 for 30 min. The catalyst was then heated to 500 °C at a rate of 10 °C/min for NH3 desorption.
The FTIR spectra of the catalysts and adsorbed species on the catalysts arising from contact with NH3 were recorded in the range 4000-650 cm−1 using a Thermo Nicolet 6700 FTIR spectrometer (Thermo Fisher, WI, USA). For the NH3 adsorption experiment, the sample was oxidized at 500 °C in a 20% O2/N2 flow (100 mL/min) for 30 min prior to NH3 chemisorption. The sample was then cooled to the desired temperature and subsequently flushed with N2 (100 mL/min) for 30 min for background collection. After that, a gas mixture containing 1000 ppm NH3 in N2 (100 mL/min) was passed through the sample for 30 min. The FTIR spectra were collected after purging with N2 for 30 min.
Figure 1 shows the NH3-SCR performance of the catalysts within the temperature range 100-350 °C. High NOx conversions were achieved over the MnOx-CeO2 catalyst at low temperatures (< 200 °C) but decreased sharply with increasing temperature. Meanwhile, its N2 selectivity was lower than 50% at 195-285 °C, accompanied by the formation of large amounts of NO2 and especially N2O. The operation window of the catalyst shifted to higher temperatures, and the N2 selectivity was improved by the addition of the acidic metal oxides in the order Nb2O5 < WO3 < MoO3. Both NbMnCe and WMnCe reached the maximal NOx conversion (98%) at 200 °C, but the latter catalyst exhibited a much broader temperature window with 86% NOx conversion at 300 °C. MoMnCe achieved higher than 90% NOx conversion at relatively higher temperatures (250-350 °C) and above 94% N2 selectivity for the whole measured temperature range.
The formation of NO2, which is important for the so-called fast-SCR reaction at low temperatures, was evaluated using the NO oxidation test. The results are shown in Fig. 2. The MnCe catalyst was deactivated for NO oxidation by the addition of the acidic metal oxides. The NO oxidation activities of the catalysts correlated well with their low-temperature SCR activities. In the case of the MnCe and NbMnCe catalysts, the NO conversion decreased at 350 °C owing to the limitation of equilibrium of NO↔NO2 redox cycle in excess O2.
Generally, the NH3 oxidation would be expected to lead to a lack of the reductant and thereby a decrease in SCR activity, especially at high temperatures. As shown in Fig. 3, the addition of an acidic metal oxide decreased the NH3 oxidation activity of the catalysts to different degrees. Modification with Nb2O5 did not obviously affect the oxidation of ammonia, while the reaction was significantly suppressed by the addition of MoO3. The addition of WO3 led to a modest inhibition effect.
Power X-ray diffraction was used to identify the phase of the catalysts as shown in Fig. 4. All the samples exhibited typical diffraction peaks of the fluorite-like structured CeO2 phase. No diffraction peaks of MnOx were found in the patterns of MnCe, NbMnCe, and WMnCe. This, as well as smaller CeO2 lattice constants than that of pure CeO2 (around 0.5412 nm), implies the replacement of Ce4+ by Mn3+ in the fluorite structure owing to their structural similarity [10, 11]. A weak characteristic peak of Mn3O4 was detected in MoMnCe, which may be owed to the buffering effect of MoO3 on the formation of MnOx-CeO2 solid solutions. The absence of the peaks assigned to MOx (M = Nb, Mo, and W) implies that these acidic metal oxides may have existed in highly dispersed or amorphous states, or were incorporated into the CeO2 lattice. The peaks of CeO2 were broadened on the addition of Nb2O5 and WO3 compared with those in the binary mixture, which is consistent with the sequence of the BET surface areas of the catalysts, i.e., WMnCe (77 m2/g) > NbMnCe (55 m2/g) > MoMnCe (36 m2/g) ≈ MnCe (35 m2/g). The shrinkage of the crystal unit cell in the Nb- and W-modified samples may have resulted from the transition of Ce3+ (0.114 nm) to Ce4+ (0.097 nm) or the incorporation of smaller Nb5+ (0.064 nm) and W6+ (0.060 nm) cations into the CeO2 lattice. Considering the relatively higher surface area of NbMnCe and WMnCe, the latter case, i.e., incorporation of the third metal during the sol-gel process, seems to be more plausible. The segregation of MnOx (Mn4+: 0.053 nm; Mn3+: 0.065 nm; Mn2+: 0.083 nm) was likely responsible for the cell expansio n of CeO2 in MoMnCe despite possible incorporation of Mo6+ (0.060 nm).
The redox properties of the catalysts were characterized by H2-TPR as shown in Fig. 5. Five reduction peaks were found in the H2-TPR curve of the MnCe catalyst. Assignment of the low-temperature (<500 °C) peaks to different MnOx species or to specific reduction steps was not straightforward because the peaks are not only related to the oxidation state but also to the crystallinity of MnOx. According to our previous study [22] and Delimaris et al.’s report [23], the first two peaks at 205 and 295 °C were related to the successive reduction of readily reducible Mnx+ species in the solid solutions or in the form of highly dispersed MnOx clusters. The next two small peaks at 360 and 450 °C corresponded to the typical two-step reduction of large MnOx crystallites. The first step involved the reduction of MnO2/Mn2O3→Mn3O4, and the second step was the reduction of Mn3O4→MnO. The oxidation state of MnOx appeared to be a mixture of MnO2 and Mn2O3 because the catalyst was calcined at 600 °C [11]. The reduction of Ce4+→Ce3+ promoted by adjacent Mnx+ ions could also occur at low temperatures, which contributed to the low-temperature reduction peaks. The broad reduction peak centered at 700 °C corresponded to the reduction of bulk oxygen of CeO2.
The H2-TPR curve of NbMnCe was similar to that of MnCe, with the reduction peaks shifted toward higher temperatures to some extent. This implied that the addition of Nb2O5 retarded the reducibility of surface oxygen on MnOx-CeO2 mixed oxides, which was similar to a previous report on the weakened reducibility of CeO2-ZrO2 mixed oxides by Nb-doping [24]. Meanwhile, Nb2O5 was difficult to be reduced in the measured temperature range. A more pronounced depressing phenomenon of WO3 was observed in the reduction curve of the WMnCe catalyst. The reduction peaks of different MnOx and Ce4+ species overlapped each other, and a distinct peak appeared at around 385 °C. As indicated by the XRD results, the segregation and sintering of Mn3O4 crystallites led to the reduction peak centered at 465 °C in the MoMnCe curve. The typical reduction peaks of MoO3, which have been reported to be located at 680 and 860 °C [25], were not observed. Consequently, the synergistic effect between CeO2 and MnOx, as well as the sintering of CeO2-based oxides, could be affected by the addition of a third metal oxide. It has been suggested that N2O formation in the NH3-SCR reaction is particularly caused by the presence of highly reactive oxygen [11, 19]. This explained the low selectivity to N2 over the MnCe catalyst. The addition of acidic metal oxides Nb2O5, WO3, and especially MoO3 decreased the reducibility of surface and lattice oxygen in the catalyst, resulting in worsened low-temperature SCR activity, enhanced high- temperature activity, and higher N2 selectivity.
NH3-TPD is a basic method used to measure the amount and strength of acid sites present on NH3-SCR catalysts from the quantity of NH3 desorbed and desorption temperature, respectively. The results are shown in Fig. 6. The total acid amount was estimated by integrating the peak area. The amount of NH3 desorbed followed the order MnCe (137 μmol/gcat) < MoMnCe (412 μmol/gcat) < WMnCe (455 μmol/gcat) < NbMnCe (490 μmol/gcat). Modification with a metal oxide significantly enhanced the acidity of the MnOx-CeO2 catalyst. If we considered the amount of NH3 desorbed per unit surface area, the order of the normalized acidity of the catalysts followed MoMnCe (11.4 μmol/(m2·gcat)) > NbMnCe (8.9 μmol/(m2·gcat)) > WMnCe (5.9 μmol/(m2·gcat)) > MnCe (3.9 μmol/(m2·gcat)). Thus, the high surface areas of NbMnCe and WMnCe contributed considerably to their uptake of NH3. Generally, there were at least two types of acid sites present on these samples, with two overlapping broad peaks at 100 (L) and 150-180 °C (H). The H/L peak area ratio was about 1.1 and 1.6-1.8 for MnCe and the modified oxides, respectively. The desorption of NH3 ended at 250 °C for MnCe, while it persisted above 300 °C for the metal oxide-modified samples. According to the NH3 adsorption IR spectra in Fig. 7, the NH3 desorbed at relatively high temperatures should be mainly assigned to that coordinated on Brönsted acid sites. In this sense, a larger amount of strong acid sites were generated on the modified catalysts than the unmodified MnCe.
Figure 7 shows the IR spectra of NH3 species adsorbed on the catalysts at room temperature for 30 min. On MnCe, bands were found at 1645 and 1290-1190 cm−1, which were assigned to the asymmetric and symmetric deformations of NH3 coordinated on Lewis acid sites on Ce4+, respectively [26, 27]. The band at 1540 cm−1 was attributed to an amide (NH2) species (scissoring mode) [11, 26]. After modification of the catalyst with the acidic metal oxides, bands were found at 1685 and 1435 cm−1, which were assigned to symmetric and asymmetric deformations of ammonium ions, respectively, resulting from NH3 adsorption on Brönsted acid sites [26, 27]. These Brönsted acid sites might have arisen from the unsaturated coordination of Ce3+ and acidic metal ions (Nb5+, W6+ and Mo6+) [27]. The band at 1435 cm−1 was extremely strong in the infrared spectrum of MoMnCe, which was consistent with the NH3-TPD result that MoO3 addition could increase the amount and strength of Brönsted acid sites on the catalyst surface. Meanwhile, the Lewis acid site-associated bands at 1645 (overlapped) and 1190 cm−1 were also observed. Many more Brönsted acid sites were formed on MoMnCe, while a large amount of Lewis acid sites were generated on the NbMnCe catalyst. These results correlated well with the above normalized amounts of acid sites per unit surface area determined by NH3-TPD. A weak band was also observed at 1060 cm−1, which was attributed to coordinatively adsorbed NH3 on different Lewis acid sites associated with the impregnated metal ions [26]. The band at 933 cm−1 was attributed to weakly adsorbed or gas phase NH3, which did not appear in the spectrum of MnCe [11, 19]. All these results demonstrated the larger amount s of acid sites and stronger acidities of the modified oxides.
Figure 8 shows the NOx desorption curves of the catalysts after pre-adsorption in NO+O2. For MnCe, NbMnCe, and WMnCe, three NO2 desorption peaks were identified in Fig. 8(a). According to the literature [28], the low-temperature peak at 95 °C always resulted from the decomposition of nitrite species (ad-NO2−), while the high-temperature peaks at 185 and 275 °C were caused by the decomposition of strongly bound nitrate species. As shown in Fig. 8(b), only a trace of NO was desorbed at low temperatures, and a small NO desorption peak appeared at around 340 °C, which was mainly attributed to the decomposition of strongly bound nitrate species [19]. The total amounts of NOx desorbed on the different catalysts followed the sequence NbMnCe (280 μmol/gcat) > WMnCe (233 μmol/gcat) > MnCe (222 μmol/gcat) > MoMnCe (35 μmol/gcat). The amount of NOx desorbed on NbMnCe and WMnCe was even larger than that on MnCe, attributed to their higher surface areas. In comparison, the high-temperature NO2 desorption peak at around 275 °C disappeared, and the total amount of NOx desorption decreased sharply on MnMoCe. This was ascribed to the low surface area and strong acidity of the catalyst. Considering the amount of NOx desorption per unit surface area, the normalized NOx adsorption capacities of the catalysts were MnCe (6.4 μmol/(m2·gcat)) > NbMnCe (5.1 μmol/(m2·gcat)) > WMnCe (3.0 μmol/(m2·gcat)) > MoMnCe (1.0 μmol/(m2·gcat)). This order reflected the reduced NOx adsorption capacity of MnOx-CeO2 after the addition of acidic metal oxides.
Figure 9 shows IR spectra of the catalysts after NO+O2 co-adsorption at room temperature for 30 min. The species observed on the different catalysts were similar but had different intensities. The gaseous or weakly adsorbed NO2 was characterized by an IR band at 1630-1620 cm−1 owing to νas (NO2) vibration [11, 27] or bridged nitrate [26]. The bands at 1545 and 1255 cm−1 on MnCe, NbMnCe, and WMnCe were attributed to the split ν3 vibration of bidentate nitrate “type I” species [26]. Bidentate nitrate “type II” species, with bands at 1530 and 1295 cm−1, became predominant on the MoMnCe catalyst [26]. Schraml-Marth et al. [29] also classified bidentate nitrates into different types on the basis of their thermal stability. It has been reported that the W addition to CeO2/TiO2 catalyst works upon not the form but the thermal stability of the adsorbed nitrate species [27].
The bands at 1450-1435 and 1025-1020 cm−1 were assigned to the ν3 and ν1 stretch vibrations of a linearly coordinated nitrite [26]. The additional band at 1195 cm−1 was assigned to anionic nitrosyl (NO−) species on CeO2, which could be readily oxidized and stored as different nitrate species in the presence of oxygen [11, 19]. It was obvious that the oxidation of NO− was much slower on MnCe than on NbMnCe and WMnCe. The NO− band was not observed on MoMnCe. Although the amount of NOx desorbed on MoMnCe was small, the corresponding IR band intensities were relatively strong, which might be related to adsorption of NOx on the surface of the segregated Mn3O4 instead of diffusion into the bulk of the catalyst.
On the basis of the above structural characterizations, the incorporation of Nb2O5 and WO3 into MnOx-CeO2 mixed oxides by sol-gel results in shrinkage of the CeO2 unit cell and an increased surface area of the catalyst. The introduction of MoO3 cannot inhibit the segregation of MnOx, and so a similar BET surface area to that of MnCe is obtained. Based on the NH3-TPD and IR spectra, the normalized NH3 adsorption of catalyst per unit surface area follows the order MoMnCe > NbMnCe > WMnCe > MnCe. This indicates that the surface acidity of the catalyst is significantly strengthened by the addition of acidic metal oxides. Especially, the amount of Brönsted acid sites is increased on the modified catalysts. However, this does not result in any improvement of the low-temperature SCR activity of catalyst, indicating that the NH3 adsorption is not expected to be the rate-determining step. Both Langmuir-Hinshelwood (L-H) and Eley-Rideal (E-R) pathways in the NO reduction over MnOx-CeO2 catalyst require the adsorption of NH3 on acid sites and formation of NH2 by partial oxidation [10, 11]. The formation of NH2 has been suggested as the key step in the mechanism of NO reduction over V-TiO2 [30]. According to the NH3 adsorption IR results, the high relative intensity of the band at 1540 cm−1 in the spectrum of MnCe indicates a strong generation of amide on this catalyst. This result correlates with the superior redox properties, NH3 oxidation activity, and low-temperature SCR activity of MnCe compared with those of the modified catalysts.
Eigenmann et al. [31] confirmed that after the initial step of NH3 adsorption and partial oxidation, nitrogen formation on MnOx-CeO2 catalysts follows an E-R type mechanism at 100-150 °C, where adsorbed NH3 reacts with NOx in the gas phase, and adsorbed NOx shows no significant reactivity. In contrast, Xu et al. [32] presented a simplified L-H mechanism for MnOx/CeO2 catalyst, in which NO molecules are adsorbed to generate NO− species, which subsequently react with the activated NH3 to form the [NH3...NO−] complex as an intermediate. Some researchers have proposed the co-existence of two different reaction routes over MnOx/Al2O3 [26], CeO2-WO3/TiO2 [27], and Ti-Fe-Mn mixed oxides [33]. They suggested that adsorbed NH3 species can either react with NO in the gas phase (E-R mechanism) or with nitrite species in the adsorbed phase (L-H mechanism) to form N2. In this study, the acidic dopants increased the amount of acid sites and enhanced the surface acidity of the catalysts, which reduced the NOx adsorption capacity and stability of nitrate species. The normalized NOx adsorption capacity of the catalysts per unit surface area was in the sequence MnCe > NbMnCe > WMnCe > MoMnCe. In this sense, the addition of acidic metal oxides may slow the reaction through an L-H mechanism [28]. This, as well as the inhibited generation of NO2 from NO oxidation, is likely to be another factor responsible for the decreased low-temperature activities of the modified catalysts. Both coordinated NH3 and adsorbed NH4+ participate in the E-R reaction pathway with increasing the reaction temperat ure. The availability of NH3 becomes critical owing to the competitive oxidation of NH3 by oxygen.
The maximal NOx conversion and the selectivity to N2 were clearly enhanced over the modified MnOx-CeO2 catalysts, and their deNOx efficiencies were improved at relatively high temperatures. It has been suggested that N2O formation occurs especially on well-ordered MnOx crystalline planes owing to the presence of highly reactive oxygen [11]. In our work, the redox properties of the catalyst were weakened by the addition of an acidic component, thereby increasing the selectivity for SCR. Meanwhile, the oxidation of NH3 was also suppressed on the modified catalysts. This, as well as enhanced NH3 adsorption on Brönsted acid sites provided by the acidic metal oxides, ensured the efficient NH3-SCR reaction at relatively high temperatures.
The combined effects of acidity and redox properties on the NH3-SCR activity of MnOx-CeO2 catalyst were investigated by introducing an acidic component using sol-gel. Incorporation of Nb2O5 and WO3 into the crystal cell of CeO2 increases the BET surface area of NbMnCe and WMnCe catalysts and does not affect the redox property severely. In contrast, failures in the incorporation of MoO3 and formation of MnOx-CeO2 solid solutions result in significant losses in the reactivity and amount of surface active sites on MoMnCe catalyst. Although the NH3 adsorption capacity is greatly increased owing to the increase in the amounts of Lewis and Brönsted acid sites, the partial oxidation of the adsorbed NH3 to amide is limited on these modified catalysts. This is suggested to be the key step in the E-R mechanism of NO reduction on MnOx-CeO2 based catalysts at low temperatures (100−150 °C). The higher acidity reduces the NOx adsorption and, with the inhibited generation of NO2, may also slow down the so-called fast SCR reaction through an L-H mechanism. However, the stable ammonium ions on Brönsted acid sites and suppressed oxidation of NH3 contribute to high SCR activity and high selectivity of MoMnCe catalyst at relatively high temperatures (250-350 °C). In this way, the temperature window of MnOx-CeO2 catalysts can be tailored by adding different acidic metal oxides to meet different specific requirements.
2014, Vol. 35 Issue (8): 1281-1288
PDF (688 KB)
2014, Vol. 35 Issue (8): 1281-1288
PDF (688 KB)