Enhanced Catalytic Performance of Hierarchical MnO x / ZSM-5 Catalyst for the Low-Temperature NH 3 -SCR

: A ZSM-5 zeolite with a hierarchical pore structure was synthesized by the desilication- recrystallization method using tetraethyl ammonium hydroxide (TEAOH) and cetyltrimethylammonium bromide (CTAB) as the desilication and structure-directing agents, respectively. The MnO x / ZSM-5 catalyst was synthesized by the ethanol dispersion method and applied for the low-temperature selective catalytic reduction of NO x with NH 3 . The results showed that NO x conversion of the hierarchical MnO x / ZSM-5 catalyst could reach 100% at about 120 ◦ C and could be maintained in the temperature range of 120–240 ◦ C with N 2 selectivity over 90%. Furthermore, the hierarchical MnO x / ZSM-5catalyst presented better SO 2 resistance performance than the traditional catalyst in the presence of 100 ppm SO 2 at 120 ◦ C. XRD, SEM, TEM, XPS, BET, NH 3 -TPD, and TG were applied to characterize the structural properties of the MnO x / ZSM-5 catalysts. These results showed that the MnO x / ZSM-5 catalyst had micropores (0.78 nm) and mesopores (3.2 nm) leading to a larger speciﬁc surface area, which improved the mass transfer of reactants and products while reducing the formation of sulfates. The better catalytic performance over hierarchical MnO x / ZSM-5 catalyst could be attributed to the higher concentration of Mn 4 + and chemisorbed oxygen species and higher surface acidity. The improved SO 2 resistance was related to the catalyst’s hierarchical pore structure.


Introduction
Selective catalytic reduction of NO x by NH 3 (NH 3 -SCR) is the most widely applied technology to remove NO x from stationary sources. The catalyst is the key factor that can determine the efficiency of the selective catalytic reduction system [1]. The temperature window of traditional commercial catalyst (V 2 O 5 -WO 3 (MoO 3 )/TiO 2 ) is 300-400 • C, and low-temperature NH 3 -SCR catalysts, which can be placed downstream of the desulfurizer and electrostatic precipitator, have attracted increasing attention in recent years [2]. Research on the low-temperature NH 3 -SCR reaction using transition metal oxides (MnO x [3], FeO x [4], CuO x [5], CeO x [6]) as the active components has become widespread. Among these metal oxides, MnO x exhibits excellent low-temperature catalytic activity, which can be attributed to the variable valence states of manganese and oxygen species [7]. However, there still remains a challenge of SO 2 poisoning for low-temperature NH 3 -SCR catalysts. The deposition of sulfite and sulfate species and sulfate of manganese oxide on the catalysts' surface lead to the inhibition of the

XRD Results
The XRD patterns of MnO x /HZ-ET (hierarchical MnO x /ZSM-5, the treated ZSM-5 was denoted as HZ-ET) and MnO x /Z-P (conventional MnO x /ZSM-5, ZSM-5 zeolite without treatment was denoted as Z-P) are shown in Figure 1. Two diffraction peaks between 2θ = 7 and 10 • and three diffraction peaks between 2θ = 22 and 25 • were observed, and these peaks were attributed to the characteristic diffraction peaks of the MFI zeolite, indicating that the MFI characteristic structure of ZSM-5 was not destroyed after desilication-recrystallization. No peaks of manganese oxide could be seen in the XRD spectra, indicating that manganese oxide in the catalyst existed either in an amorphous or highly dispersed state. TEM images would be used to further elucidate the structure of the MnO x . The relative crystallinity results of MnO x /HZ-ET and MnO x /Z-P are shown in Table 1. Compared with the parent ZSM-5 (Z-P), the lower relative crystallinity (R C ) of the MnO x /HZ-ET was attributed to the desilication, which destroyed the crystal structure [16] and the interaction between the MnO x and the zeolite [17]. The R C of MnO x /HZ-ET was higher than MnO x /Z-P because of recrystallization, which indicated that the framework structure of MnO x /HZ-ET was more complete.               Combined with the XRD analysis, the results of TEM images showed that manganese oxide was not amorphous, but existed in various forms of crystallization in highly dispersed states. The (110) planes of MnO 2 in Figure 3e was considered as the most active crystal plane in the NH 3 -SCR reaction [18] and provided a large number of active sites and oxygen vacancies. This was one of the reasons why the MnO x /HZ-ET catalyst showed better low-temperature SCR activity than MnO x /Z-P.

SEM and TEM Results
Catalysts 2020, 10, x FOR PEER REVIEW 4 of 12 in various forms of crystallization in highly dispersed states. The (110) planes of MnO2 in Figure 3e was considered as the most active crystal plane in the NH3-SCR reaction [18] and provided a large number of active sites and oxygen vacancies.. This was one of the reasons why the MnOx/HZ-ET catalyst showed better low-temperature SCR activity than MnOx/Z-P.  Figure 4 shows the nitrogen adsorption-desorption isotherms and the pore size distribution curves of MnOx/Z-P and MnOx/HZ-ET catalysts. Figure 4a shows that MnOx/Z-P exhibited type I isotherms with no distinct hysteresis loop, which indicated that MnOx/Z-P was a typical microporous material, where MnOx/HZ-ET displayed typical type I isotherms at low relative pressure (P/P0 < 0.01) and type IV isotherms with a clear hysteresis loop at higher relative pressure (P/P0 > 0.4), demonstrating the co-existence of micropores and mesopores created through the desilication-recrystallization processes. Figure 4b shows that the micropore size distributions of MnOx/HZ-ET catalyst centered around 0.78 nm, which was larger than that of MnOx/Z-P (around 0.73 nm). The enlargement of the micropores in MnOx/HZ-ET may be due to the destruction of pore walls by treatment with TEAOH [19]. The pore size distribution of MnOx/Z-P ( Figure 4b) showed no obvious peaks in the range of 2-20 nm, and MnOx/HZ-ET showed a narrow and intense peak at 3.2 nm. In addition, MnOx/HZ-ET showed broad peaks between 4 and 15 nm, which possibly indicated that the intracrystalline mesopores were caused by desilication and existed in an irregular state [16]; this result was consistent with the TEM images.   Figure 4 shows the nitrogen adsorption-desorption isotherms and the pore size distribution curves of MnO x /Z-P and MnO x /HZ-ET catalysts. Figure 4a shows that MnO x /Z-P exhibited type I isotherms with no distinct hysteresis loop, which indicated that MnO x /Z-P was a typical microporous material, where MnO x /HZ-ET displayed typical type I isotherms at low relative pressure (P/P 0 < 0.01) and type IV isotherms with a clear hysteresis loop at higher relative pressure (P/P 0 > 0.4), demonstrating the co-existence of micropores and mesopores created through the desilication-recrystallization processes. Figure 4b shows that the micropore size distributions of MnO x /HZ-ET catalyst centered around 0.78 nm, which was larger than that of MnO x /Z-P (around 0.73 nm). The enlargement of the micropores in MnO x /HZ-ET may be due to the destruction of pore walls by treatment with TEAOH [19]. The pore size distribution of MnO x /Z-P (Figure 4b) showed no obvious peaks in the range of 2-20 nm, and MnO x /HZ-ET showed a narrow and intense peak at 3.2 nm. In addition, MnO x /HZ-ET showed broad peaks between 4 and 15 nm, which possibly indicated that the intracrystalline mesopores were caused by desilication and existed in an irregular state [16]; this result was consistent with the TEM images.  Table 2 shows the corresponding results of structural properties of MnOx/Z-P and MnOx/HZ-ET catalysts. The specific surface areas and the external surface areas of MnOx/HZ-ET were 328 m 2 /g and 124 m 2 /g, which was higher than MnOx/Z-P (247 m 2 /g and 12 m 2 /g, respectively). MnOx/HZ-ET contained more mesopores and less micropores than MnOx/Z-P, which was attributed to the transformation of part of the microporous phase into the mesoporous phase. The larger specific surface area and pore size within hierarchical MnOx/HZ-ET improved the mass transfer of reactants and products during the NH3-SCR reaction, thus improving its catalytic performance compared to MnOx/Z-P, which only had microporous structures.

XPS Results
The surface atomic composition was analyzed by XPS. Figure 5 shows the XPS spectra of Mn 2p and O 1s for the MnOx/Z-P and MnOx/HZ-ET catalysts. The XPS spectra of Mn 2p3/2 could be fit into three characteristic peaks ascribed to Mn 2+ at 640.4 eV, Mn 3+ at 642.0 eV, and Mn 4+ at 644.1 eV. The XPS spectra of O 1s could be divided into two characteristic peaks ascribed to lattice oxygen species (Oα), around 529.7 eV, and chemisorbed oxygen species (Oβ), around 531.5 eV [20]. The corresponding results are summarized in Table 3. It was clear that the molar concentration of manganese on the surface of MnOx/HZ-ET (9.62%) was higher than MnOx/Z-P (8.47%), and the relative surface concentration ratios of Mn 4+ (Mn 4+ /Mn) of MnOx/HZ-ET(49.09%) were much higher than MnOx/Z-P (36.36%). It was reported that Mn 4+ could promote the oxidation of NO to NO2, and higher concentration ratios of Mn 4+ species were beneficial for the improvement of their redox cycle, thus promoting the activity in the NH3-SCR reaction at low temperatures [21]. In addition, the percentage of chemisorbed oxygen species Oβ in MnOx/HZ-ET (87.09%) was much higher than MnOx/Z-P (75.41%). The chemisorbed oxygen species Oβ was reported to be more active than Oα because of their higher mobility. Higher percentage of Oβ was beneficial for the oxidation of NO to NO2, enhancing the low-temperature activity through the "fast SCR" reaction [22]. Therefore, the excellent low-temperature SCR activity of MnOx/HZ-ET was attributed to the higher percentage of active Mn 4+ and chemisorbed oxygen species.  Table 2 shows the corresponding results of structural properties of MnO x /Z-P and MnO x /HZ-ET catalysts. The specific surface areas and the external surface areas of MnO x /HZ-ET were 328 m 2 /g and 124 m 2 /g, which was higher than MnO x /Z-P (247 m 2 /g and 12 m 2 /g, respectively). MnO x /HZ-ET contained more mesopores and less micropores than MnO x /Z-P, which was attributed to the transformation of part of the microporous phase into the mesoporous phase. The larger specific surface area and pore size within hierarchical MnO x /HZ-ET improved the mass transfer of reactants and products during the NH 3 -SCR reaction, thus improving its catalytic performance compared to MnO x /Z-P, which only had microporous structures.

XPS Results
The surface atomic composition was analyzed by XPS. Figure 5 shows the XPS spectra of Mn 2p and O 1s for the MnO x /Z-P and MnO x /HZ-ET catalysts. The XPS spectra of Mn 2p 3/2 could be fit into three characteristic peaks ascribed to Mn 2+ at 640.4 eV, Mn 3+ at 642.0 eV, and Mn 4+ at 644.1 eV. The XPS spectra of O 1s could be divided into two characteristic peaks ascribed to lattice oxygen species (O α ), around 529.7 eV, and chemisorbed oxygen species (O β ), around 531.5 eV [20]. The corresponding results are summarized in Table 3. It was clear that the molar concentration of manganese on the surface of MnO x /HZ-ET (9.62%) was higher than MnO x /Z-P (8.47%), and the relative surface concentration ratios of Mn 4+ (Mn 4+ /Mn) of MnO x /HZ-ET(49.09%) were much higher than MnO x /Z-P (36.36%). It was reported that Mn 4+ could promote the oxidation of NO to NO 2 , and higher concentration ratios of Mn 4+ species were beneficial for the improvement of their redox cycle, thus promoting the activity in the NH 3 -SCR reaction at low temperatures [21]. In addition, the percentage of chemisorbed oxygen species O β in MnO x /HZ-ET (87.09%) was much higher than MnO x /Z-P (75.41%). The chemisorbed oxygen species O β was reported to be more active than O α because of their higher mobility. Higher percentage of O β was beneficial for the oxidation of NO to NO 2 , enhancing the low-temperature activity through the "fast SCR" reaction [22]. Therefore, the excellent low-temperature SCR activity of MnO x /HZ-ET was attributed to the higher percentage of active Mn 4+ and chemisorbed oxygen species. Intensity (a.u.) Binding Energy (eV)

NH3-TPD Results
The surface acidity properties of the catalysts were analyzed by the NH3-TPD technique. Figure  6 shows that the NH3 desorption profiles of MnOx/HZ-ET and MnOx/Z-P exhibited two distinct desorption peaks. The desorption peaks below 350 °C were associated with the weakly acidic sites, while the desorption peaks ranging from 350 to 550 °C were associated with the strongly acidic sites [23]. The quantified results of NH3-TPD are summarized in Table 4, and the area of the peak was proportional to the amount of acid. It could be found that MnOx/HZ-ET showed more total acid amount than MnOx/Z-P. Although the weak acid amount of MnOx/Z-P was higher than MnOx/HZ-ET, the strong acid and total acid amount of MnOx/HZ-ET were both higher than MnOx/Z-P, which may be ascribed to the higher specific surface area of the MnOx/HZ-ET catalyst. It could be concluded that MnOx/HZ-ET with a hierarchical pore structure could provide more strong surface acidity sites, which would be beneficial for the adsorption and activation of NH3, resulting in increased low-temperature NH3-SCR performance.

NH 3 -TPD Results
The surface acidity properties of the catalysts were analyzed by the NH 3 -TPD technique. Figure 6 shows that the NH 3 desorption profiles of MnO x /HZ-ET and MnO x /Z-P exhibited two distinct desorption peaks. The desorption peaks below 350 • C were associated with the weakly acidic sites, while the desorption peaks ranging from 350 to 550 • C were associated with the strongly acidic sites [23]. The quantified results of NH 3 -TPD are summarized in Table 4, and the area of the peak was proportional to the amount of acid. It could be found that MnO x /HZ-ET showed more total acid amount than MnO x /Z-P. Although the weak acid amount of MnO x /Z-P was higher than MnO x /HZ-ET, the strong acid and total acid amount of MnO x /HZ-ET were both higher than MnO x /Z-P, which may be ascribed to the higher specific surface area of the MnO x /HZ-ET catalyst. It could be concluded that MnO x /HZ-ET with a hierarchical pore structure could provide more strong surface acidity sites, which would be beneficial for the adsorption and activation of NH 3 , resulting in increased low-temperature NH 3 -SCR performance.

NH3-TPD Results
The surface acidity properties of the catalysts were analyzed by the NH3-TPD technique. Figure  6 shows that the NH3 desorption profiles of MnOx/HZ-ET and MnOx/Z-P exhibited two distinct desorption peaks. The desorption peaks below 350 °C were associated with the weakly acidic sites, while the desorption peaks ranging from 350 to 550 °C were associated with the strongly acidic sites [23]. The quantified results of NH3-TPD are summarized in Table 4, and the area of the peak was proportional to the amount of acid. It could be found that MnOx/HZ-ET showed more total acid amount than MnOx/Z-P. Although the weak acid amount of MnOx/Z-P was higher than MnOx/HZ-ET, the strong acid and total acid amount of MnOx/HZ-ET were both higher than MnOx/Z-P, which may be ascribed to the higher specific surface area of the MnOx/HZ-ET catalyst. It could be concluded that MnOx/HZ-ET with a hierarchical pore structure could provide more strong surface acidity sites, which would be beneficial for the adsorption and activation of NH3, resulting in increased low-temperature NH3-SCR performance.

Sample
Area of peak Total area of peaks Peak 1 Peak 2 Figure 6. NH 3 -TPD of the catalysts.

SCR Performance
Low-temperature NH 3 -SCR activities and N 2 selectivities of MnO x /HZ-ET and MnO x /Z-P were tested over the temperature range of 80-240 • C, and the results of the NO x conversion are shown in Figure 7. The MnO x /Z-P catalyst showed low NO x conversion in the temperature range of 80-180 • C, and 99% NO x conversion was obtained around 180 • C. Meanwhile, the N 2 selectivity of MnO x /Z-P was lower than 90% over the whole temperature range. Compared with MnO x /Z-P, MnO x /HZ-ET showed significantly higher NO x conversion over the temperature range of 80-120 • C. When the temperature reached 120 • C, nearly 100% NO x conversion could be obtained with a broader operating temperature window (120-240 • C). Over 90% N 2 , selectivity could be maintained for the MnO x /HZ-ET catalyst throughout the entire temperature range. Conventional microporous catalysts were reported to have several drawbacks in the SCR reaction, such as diffusion limitations of the reactants and products [24]. Therefore, it was reasonable to deduce that the mass transfer of reactants and products could be enhanced with the existence of mesopores at low temperatures, and as a result of this, better low-temperature NH 3 -SCR performance could be obtained using the hierarchical MnO x /HZ-ET catalyst.

SCR Performance
Low-temperature NH3-SCR activities and N2 selectivities of MnOx/HZ-ET and MnOx/Z-P were tested over the temperature range of 80-240 °C, and the results of the NOx conversion are shown in Figure 7. The MnOx/Z-P catalyst showed low NOx conversion in the temperature range of 80-180 °C, and 99% NOx conversion was obtained around 180 °C. Meanwhile, the N2 selectivity of MnOx/Z-P was lower than 90% over the whole temperature range. Compared with MnOx/Z-P, MnOx/HZ-ET showed significantly higher NOx conversion over the temperature range of 80-120 °C. When the temperature reached 120 °C, nearly 100% NOx conversion could be obtained with a broader operating temperature window (120-240 °C). Over 90% N2, selectivity could be maintained for the MnOx/HZ-ET catalyst throughout the entire temperature range. Conventional microporous catalysts were reported to have several drawbacks in the SCR reaction, such as diffusion limitations of the reactants and products [24]. Therefore, it was reasonable to deduce that the mass transfer of reactants and products could be enhanced with the existence of mesopores at low temperatures, and as a result of this, better low-temperature NH3-SCR performance could be obtained using the hierarchical MnOx/HZ-ET catalyst.  Figure 8 shows the NOx conversion for the MnOx/HZ-ET and MnOx/Z-P catalysts in the presence of 100 ppm SO2 at 120 °C and 180 °C, respectively. The results showed that NOx conversion noticeably decreased with the addition of SO2 into the feed gas. The SCR activity of the MnOx/Z-P catalyst decreased from 100% to 15% after 60 min of SO2 addition. The NOx conversion of MnOx/HZ-ET decreased from 100% to 60% after the SO2 was added for 1 h. The NOx conversion could not be recovered over the two catalysts when SO2 was absent from the reaction mixture, which indicated that the deactivation was irreversible. (NH4)2SO4 or NH4HSO4 may be formed during the reaction when SO2 was added into the reaction atmospheres. These species could only decompose over 300 °C and may block the zeolite channels, leading to reduced catalytic activities. At the same time, catalysts with a larger pore size could reduce the formation of sulfate species [25]. Comparing the SO2 tolerance of MnOx/HZ-ET and MnOx/Z-P, the hierarchical pore structure MnOx/HZ-ET catalyst could provide larger specific surface area and larger pore size, which was beneficial for its SO2 resistance.  Figure 8 shows the NO x conversion for the MnO x /HZ-ET and MnO x /Z-P catalysts in the presence of 100 ppm SO 2 at 120 • C and 180 • C, respectively. The results showed that NO x conversion noticeably decreased with the addition of SO 2 into the feed gas. The SCR activity of the MnO x /Z-P catalyst decreased from 100% to 15% after 60 min of SO 2 addition. The NO x conversion of MnO x /HZ-ET decreased from 100% to 60% after the SO 2 was added for 1 h. The NO x conversion could not be recovered over the two catalysts when SO 2 was absent from the reaction mixture, which indicated that the deactivation was irreversible. (NH 4 ) 2 SO 4 or NH 4 HSO 4 may be formed during the reaction when SO 2 was added into the reaction atmospheres. These species could only decompose over 300 • C and may block the zeolite channels, leading to reduced catalytic activities. At the same time, catalysts with a larger pore size could reduce the formation of sulfate species [25]. Comparing the SO 2 tolerance of MnO x /HZ-ET and MnO x /Z-P, the hierarchical pore structure MnO x /HZ-ET catalyst could provide larger specific surface area and larger pore size, which was beneficial for its SO 2 resistance. XRD and TG were applied to identity the formation of (NH4)2SO4 or NH4HSO4. Figure 9 shows the XRD patterns of poisoned MnOx/HZ-ET and MnOx/Z-P catalysts. The peak at 2θ = 33° observed over the used MnOx/Z-P could be attributed to the phase of formed NH4HSO4. No obvious peaks of NH4HSO4 could be observed in the XRD pattern of used MnOx/HZ-ET, which may be because the NH4HSO4 formed existed in amorphous species or was below the detection limit. Figure 10 presents the TG curves of the poisoned MnOx/HZ-ET and MnOx/Z-P catalysts. The first weight loss emerging at 80-110 °C could be due to the evaporation of water in the catalysts. Two other weight losses at about 250 °C and 350 °C were close to the decomposition temperature of (NH4)2SO4 or NH4HSO4 [8]. For MnOx/HZ-ET, the intensities of two weight losses was lower than that of MnOx/Z-P. The XRD and TG results showed that the MnOx/HZ-ET with a larger pore size could reduce the deposition of (NH4)2SO4 or NH4HSO4 on the catalyst surface during the NH3-SCR reaction with SO2, which was one of the reasons for the better SO2 resistance of the MnOx/HZ-ET catalyst.  XRD and TG were applied to identity the formation of (NH 4 ) 2 SO 4 or NH 4 HSO 4 . Figure 9 shows the XRD patterns of poisoned MnO x /HZ-ET and MnO x /Z-P catalysts. The peak at 2θ = 33 • observed over the used MnO x /Z-P could be attributed to the phase of formed NH 4 HSO 4 . No obvious peaks of NH 4 HSO 4 could be observed in the XRD pattern of used MnO x /HZ-ET, which may be because the NH 4 HSO 4 formed existed in amorphous species or was below the detection limit. Figure 10 presents the TG curves of the poisoned MnO x /HZ-ET and MnO x /Z-P catalysts. The first weight loss emerging at 80-110 • C could be due to the evaporation of water in the catalysts. Two other weight losses at about 250 • C and 350 • C were close to the decomposition temperature of (NH 4 ) 2 SO 4 or NH 4 HSO 4 [8].

Effect of SO 2 on SCR Catalytic Activity
For MnO x /HZ-ET, the intensities of two weight losses was lower than that of MnO x /Z-P. The XRD and TG results showed that the MnO x /HZ-ET with a larger pore size could reduce the deposition of (NH 4 ) 2 SO 4 or NH 4 HSO 4 on the catalyst surface during the NH 3 -SCR reaction with SO 2 , which was one of the reasons for the better SO 2 resistance of the MnO x /HZ-ET catalyst. XRD and TG were applied to identity the formation of (NH4)2SO4 or NH4HSO4. Figure 9 shows the XRD patterns of poisoned MnOx/HZ-ET and MnOx/Z-P catalysts. The peak at 2θ = 33° observed over the used MnOx/Z-P could be attributed to the phase of formed NH4HSO4. No obvious peaks of NH4HSO4 could be observed in the XRD pattern of used MnOx/HZ-ET, which may be because the NH4HSO4 formed existed in amorphous species or was below the detection limit. Figure 10 presents the TG curves of the poisoned MnOx/HZ-ET and MnOx/Z-P catalysts. The first weight loss emerging at 80-110 °C could be due to the evaporation of water in the catalysts. Two other weight losses at about 250 °C and 350 °C were close to the decomposition temperature of (NH4)2SO4 or NH4HSO4 [8]. For MnOx/HZ-ET, the intensities of two weight losses was lower than that of MnOx/Z-P. The XRD and TG results showed that the MnOx/HZ-ET with a larger pore size could reduce the deposition of (NH4)2SO4 or NH4HSO4 on the catalyst surface during the NH3-SCR reaction with SO2, which was one of the reasons for the better SO2 resistance of the MnOx/HZ-ET catalyst.

Catalyst Preparation
The hierarchical ZSM-5 was prepared from commercial MFI-type zeolites by sequential desilication-recrystallization based on previous studies [16,19,26]. The parent zeolite in this study was a commercial H form ZSM-5 with a Si/Al mass ratio = 38 (XFNANO Company, Nanjing, China). Typically, three grams of ZSM-5 zeolite and 1 g cetyltrimethylammonium bromide (CTAB, Aladdin, Shanghai, China) were dispersed in 30 mL of 1 M tetraethyl ammonium hydroxide (TEAOH, Aladdin, Shanghai, China), and the mixture was stirred at ambient temperature. The solution was transferred into a Teflon-lined autoclave and treated at 150 °C for 24 h. The product was washed with distilled water, filtered, dried, and calcined at 550 °C for 4 h to remove the templates. The treated sample was denoted as HZ-ET, and for comparison, the ZSM-5 zeolite without treatment was denoted as Z-P.
The catalysts were synthesized by the ethanol dispersion method, and manganese nitrate (50 wt.% in H2O, Aladdin, Shanghai, China) was used as a precursor. Typically, one-point-nine-five grams of 50 wt.% Mn(NO3)2 (Mn loading = 15 wt.%) were dissolved in 75 mL ethanol under stirring at ambient temperature. Subsequently, two grams of HZ-ET or Z-P powder were added and stirred. Then, the solution was treated with ultrasound for 0.5 h and stirred continuously at 85 °C until the solvent was completely evaporated. The products were dried at 80 °C and calcined at 400 °C for 3 h. The catalyst was denoted as MnOx/HZ-ET. For comparison, MnOx/Z-P was also prepared according to the same method.

Low-Temperature NH3-SCR Activity Measurements
The NH3-SCR activity tests were carried out in a fixed-bed quartz glass reactor. Five-hundred milligrams of 40-60 mesh catalysts were used in the test. The reaction temperature was increased from room temperature to 240 °C with a heating rate of 5 °C/min with an isotherm step of 20 °C. The gas was composed of 800 ppm NH3, 800 ppm NO, 5.0 vol% O2, and 100 ppm SO2 (when added), balanced by Ar. The flow rate was 600 mL/min with a gas hourly space velocity (GHSV) of 40,000 h −1 . The concentrations of NO/NO2 were measured by a NO-NO2-NOx analyzer (Thermal Scientific, model 42i-HL, Waltham, USA), and N2 gas chromatography was used to analyze N2-selectivity. The NOx removal efficiency and the N2 selectivity are calculated as follows:

Catalyst Preparation
The hierarchical ZSM-5 was prepared from commercial MFI-type zeolites by sequential desilication-recrystallization based on previous studies [16,19,26]. The parent zeolite in this study was a commercial H form ZSM-5 with a Si/Al mass ratio = 38 (XFNANO Company, Nanjing, China). Typically, three grams of ZSM-5 zeolite and 1 g cetyltrimethylammonium bromide (CTAB, Aladdin, Shanghai, China) were dispersed in 30 mL of 1 M tetraethyl ammonium hydroxide (TEAOH, Aladdin, Shanghai, China), and the mixture was stirred at ambient temperature. The solution was transferred into a Teflon-lined autoclave and treated at 150 • C for 24 h. The product was washed with distilled water, filtered, dried, and calcined at 550 • C for 4 h to remove the templates. The treated sample was denoted as HZ-ET, and for comparison, the ZSM-5 zeolite without treatment was denoted as Z-P.
The catalysts were synthesized by the ethanol dispersion method, and manganese nitrate (50 wt.% in H 2 O, Aladdin, Shanghai, China) was used as a precursor. Typically, one-point-nine-five grams of 50 wt.% Mn(NO 3 ) 2 (Mn loading = 15 wt.%) were dissolved in 75 mL ethanol under stirring at ambient temperature. Subsequently, two grams of HZ-ET or Z-P powder were added and stirred. Then, the solution was treated with ultrasound for 0.5 h and stirred continuously at 85 • C until the solvent was completely evaporated. The products were dried at 80 • C and calcined at 400 • C for 3 h. The catalyst was denoted as MnO x /HZ-ET. For comparison, MnO x /Z-P was also prepared according to the same method.

Low-Temperature NH 3 -SCR Activity Measurements
The NH 3 -SCR activity tests were carried out in a fixed-bed quartz glass reactor. Five-hundred milligrams of 40-60 mesh catalysts were used in the test. The reaction temperature was increased from room temperature to 240 • C with a heating rate of 5 • C/min with an isotherm step of 20 • C. The gas was composed of 800 ppm NH 3 , 800 ppm NO, 5.0 vol% O 2 , and 100 ppm SO 2 (when added), balanced by Ar. The flow rate was 600 mL/min with a gas hourly space velocity (GHSV) of 40,000 h −1 . The concentrations of NO/NO 2 were measured by a NO-NO 2 -NO x analyzer (Thermal Scientific, model 42i-HL, Waltham, USA), and N 2 gas chromatography was used to analyze N 2 -selectivity. The NO x removal efficiency and the N 2 selectivity are calculated as follows:

Characterization
X-ray diffraction (XRD) patterns were recorded on a D8 Advance diffractometer (Bruker, Karlsruhe, Germany) with Cu Kα radiation. Chemical states of all elements were analyzed by X-ray photoelectron spectroscopy (XPS, Axis Ultra DLD, Kratos, U.K.) with Al Kα radiation (hυ = 1253.6 eV). NH 3 -TPD analysis was performed on Tp 5080 (Xianquan Industrial and Trading Co., Ltd, Tianjin, China). A scanning electron microscope (SEM) was carried out on ZEISS Merlin instrument (Carl Zeiss AG, Jena, Germany). Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) were performed on TECNAI G2 F20 (FEI, Hillsboro, OR, USA). N 2 physisorption was performed at −196 • C using a surface analyzer (Micomeritics, ASAP 2020, Norcross, USA), and all samples were degassed in a vacuum at 30 • C for 6 h prior to measurements. The Brunauer-Emmett-Teller (BET) method was used to calculate the specific surface area. The micropore surface area and micropore volume were evaluated using a t-plot. The Barrett-Joyner-Halenda (BJH) method was used to determine the pore size distributions from desorption branches. Thermogravimetric analysis (TG) was carried out in a static N2 atmosphere using a TGA/DSC 3+instrument (Mettler, Zurich, Switzerland). Fifteen milligrams of each sample were analyzed between 30 and 800 • C at a rate of 10 • C/min.

Conclusions
A MnO x /HZ-ET catalyst, which possessed a hierarchical pore structure, was successfully prepared through desilication-recrystallization and ethanol dispersion methods. Compared with MnO x /Z-P prepared using traditional microporous ZSM-5 as a support, MnO x /HZ-ET showed better low-temperature NH 3 -SCR activity with nearly 100% NO x conversion over a broad temperature window from 120 • C to 240 • C and above 90% N 2 selectivity throughout the entire temperature range. The low temperature NH 3 -SCR performance of MnO x /HZ-ET could be attributed to the hierarchical pore structure, higher concentration of chemisorbed oxygen and Mn 4+ species, as well as appropriate acid strengths and amounts. The MnO x /HZ-ET catalyst also showed better SO 2 tolerance than the MnO x /Z-P, and 60% NO x conversion could be maintained after the SO 2 was added for 1 h, which could be related to its larger specific surface area and larger pore size, which may reduce the deposition of ammonium sulfate.