Promoting Effect of Mn on In Situ Synthesized Cu-SSZ-13 for NH₃-SCR

Abstract

The effect of Mn impregnation on the NH₃-SCR (selective catalytic reduction of NO_x by NH₃) activity of in situ synthesized Cu-SSZ-13 was investigated in this work. It was found that Mn addition could efficiently improve the low-temperature activity of Cu-SSZ-13. The optimal amount of Mn was 5 wt.%, and NO_x conversion was improved by more than 20% over a temperature range of 120 °C to 150 °C. SEM (scanning electron microscopy), XRD (X-ray diffraction), N₂ adsorption-desorption, H₂-TPR (temperature programmed reduction of H₂), NH₃-TPD (temperature programmed desorption of NH₃) and in situ DRIFTS (diffuse reflectance infrared Fourier transform spectroscopy) experiments were conducted to investigate the changes in the zeolite structure, active sites, acid sites and reaction mechanism. The impregnated MnO_x species caused a decline in the crystallinity of Cu-SSZ-13 but markedly improved the redox ability. Nitrate and nitrite species were observed in the Mn-modified Cu-SSZ-13, and the formation of these species was thought to cause the observed increase in low-temperature NH₃-SCR activity. The results show that the addition of Mn is a promising method for promoting the low-temperature catalytic activity of Cu-SSZ-13.

1. Introduction

The emission of nitrogen oxides (NO_x) from heavy-duty diesel vehicles makes a large contribution to atmospheric pollution [1,2,3]. The selective catalytic reduction of NO_x by NH₃ (NH₃-SCR) has proven to be one of the most efficient technologies for NO_x abatement from heavy-duty diesel vehicles [4,5,6,7,8]. As an NH₃-SCR catalyst, Cu-SSZ-13 zeolite has, with its small-pore structure, attracted much attention in recent years [9,10,11]. NH₃-solvated Cu ions, which are weakly tethered to framework Al and regarded as mobile active sites, make the NH₃-SCR reaction occur in a near-homogeneous fashion in a limited region [12]. The pore size of Cu-SSZ-13 is around 3.8 Å, which screens large molecules such as HCs from entering the zeolite cavities and also restricts the out-diffusion of Al(OH)₃ species formed under hydrothermal conditions [13]. Therefore, the special pore structure endows Cu-SSZ-13 with excellent NH₃-SCR activity, N₂ selectivity, HCs resistance and hydrothermal stability.

Ren et al. invented an in situ synthesis method for Cu-SSZ-13 that uses Cu-tetraethylenepentamine (Cu-TEPA) as a template to directly introduce Cu into the precursor gel [14,15]. Compared with the conventional ion-exchange method, this in situ synthesis method avoids the complicated ion-exchange process and reduces the production of wastewater, which is beneficial for practical application. Furthermore, Xie et al. and Shan et al. optimized the structure of and copper species in in situ synthesized Cu-SSZ-13 by HNO₃ and NH₄NO₃ after treatment, leading to marked increases in the catalytic activity and hydrothermal stability [16,17]. Nevertheless, the low-temperature activity (<180 °C) of in situ synthesized Cu-SSZ-13 is still low and needs to be improved [18,19,20].

Numerous studies showed that, at temperatures below 180 °C, Mn-based catalysts were very active in the NH₃-SCR reaction [21,22,23,24,25,26,27]. For instance, over 90% NO_x conversion could be obtained over MnO_x-CeO₂ complex oxide catalysts at 150 °C [21,28]. Tang et al. doped a series of metals on Mn-based catalysts, and they found that the Mn-Fe-Ce catalyst performed the best. A 95% NO_x conversion could be obtained, and it also had an excellent SO₂ and H₂O tolerance [27]. Ye et al. synthesized a series of Mn/SSZ-13 catalysts with various Mn loadings, and over 95% NO_x conversion was obtained at 150 °C for the best sample [29]. Meanwhile, Mn-based bimetallic zeolites also showed excellent low-temperature NH₃-SCR activity. Kim et al. added Mn into the Fe-ZSM-5 catalyst, and the catalytic performance was improved [30]. The NO_x conversion of Cu-SSZ-39 was also raised over the whole temperature range after the addition of Mn [31]. Song et al. synthesized Cu-Mn-SSZ-13 catalysts by a coexchange method, and both the catalytic activity and hydrothermal stability were improved with the appropriate loading amount [32]. It seems that the addition of Mn has positive effects on the NH₃-SCR activity of Cu-SSZ-13. To meet the demands of a practical application, the low-temperature catalytic activity of in situ synthesized Cu-SSZ-13, which is one of the most promising catalysts utilized in heavy-duty diesel vehicles, should also be of concern.

In this study, a series of Mn/Cu-SSZ-13 catalysts with various Mn loadings were prepared by impregnating manganese acetates into in situ synthesized Cu-SSZ-13. The results showed that the addition of Mn could improve the catalytic performance of Cu-SSZ-13 catalysts. The effects of Mn impregnation on the zeolite framework and morphology of Cu-SSZ-13 were characterized by XRD and SEM. H₂-TPR and NH₃-TPD were utilized to assess the redox ability and acidity of Cu-SSZ-13 and Mn/Cu-SSZ-13. In addition, a series of in situ DRIFTS experiments were carried out to investigate the NH₃-SCR mechanism over the prepared catalysts.

2. Results and Discussion

2.1. NH₃-SCR Performance

The standard NH₃-SCR catalytic activity profiles of all the prepared catalysts are presented in Figure 1. As can be seen, all the catalysts showed a broad window of over 90% NO_x conversion from around 160 °C to 330 °C. As can be seen, after impregnation with Mn, all catalysts presented a better catalytic activity below 160 °C when compared with Cu-SSZ-13. Among all the catalysts, 5% Mn/Cu-SSZ-13 showed the best catalytic activity in the low temperature range, and nearly 95% NO_x conversion was obtained around 140 °C. With a further increase in the Mn content, however, the low-temperature NO_x conversion began to decrease when compared with 5% Mn/Cu-SSZ-13. To analyze the selectivity of the prepared catalysts, the N₂O concentrations in the effluents from the catalysts are summarized in Figure S1. More N₂O was produced with an increase of the Mn content, indicating that the addition of Mn would decrease the N₂ selectivity of the catalysts. Nevertheless, the N₂O formation was under 20 ppm for Mn/Cu-SSZ-13 with an Mn content below 7%. The accurate control of the Mn content was of vital importance to the catalytic performance, and 5% Mn/Cu-SSZ-13 performed the best among all the prepared catalysts. The following sections will compare the characterization results between 5% Mn/Cu-SSZ-13 and Cu-SSZ-13; the promotion effect of Mn on Cu-SSZ-13 was also investigated.

2.2. Characterization of Catalysts

The SEM images illustrate the morphology of Cu-SSZ-13 (Figure 2a) and 5% Mn/Cu-SSZ-13 (Figure 2b). As can be seen, in both cases the catalyst grains had a cubic shape with a crystallite size of around 300–500 nm. No discernable destruction of the grains could be observed after 5% Mn was impregnated into Cu-SSZ-13, indicating that the morphology of Cu-SSZ-13 was not affected by doping with Mn. Furthermore, the N₂ adsorption-desorption experimental results showed that the surface area and pore volume were 613 m²/g and 0.212 cm³/g for Cu-SSZ-13, and 596 m²/g and 0.206 cm³/g for 5% Mn/Cu-SSZ-13. Compared with Cu-SSZ-13, there was only a slight decrease in the surface area and pore volume for 5% Mn/Cu-SSZ-13, indicating that the channel structure was not affected by Mn impregnation. In addition, the SEM images of catalysts with other Mn contents are presented in Figure S2. The morphologies of these catalysts showed little change after Mn impregnation, except for 14% Mn/Cu-SSZ-13, which showed some degradation of the structure. In summary, when the Mn content was below 10%, the MnO_x species over Cu-SSZ-13 were highly dispersed and did not affect the channel structure of Cu-SSZ-13.

Furthermore, EDX was utilized to detect the composition of atoms in Cu-SSZ-13 and 5% Cu-SSZ-13, and the results are presented in Figure S3. Ten spots were scanned for each catalyst, and the atomic composition was calculated as the average value of the 10 spots. As can be seen, there were fewer O, Al, Si and Cu atoms in 5% Mn/Cu-SSZ-13 than in Cu-SSZ-13, which was due to the addition of Mn. Meanwhile, EDS could only detect the atoms on the surface of the catalysts. 5% Mn was added into the catalysts; however, only 1.8% Mn was detected. These results indicated that more Mn atoms were located in the pores of Cu-SSZ-13, which was in accordance with the results of the SEM images.

The XRD patterns of Cu-SSZ-13 and 5% Mn/Cu-SSZ-13 are presented in Figure 3. Both catalysts showed the typical diffraction peaks of the CHA structure, indicating that the addition of Mn did not change the zeolite structure of CHA. Meanwhile, no peaks resulting from CuO_x species could be observed for either Cu-SSZ-13 or 5% Mn/Cu-SSZ-13. Compared with Cu-SSZ-13, no new diffraction peaks emerged in the pattern of 5% Mn/Cu-SSZ-13, which meant that the MnO_x species were highly dispersed. This result was in accordance with the SEM images, which showed no evidence of MnO_x species accumulation. The XRD patterns of all the prepared catalysts are presented in Figure S4. With the increase in the Mn content, the crystallinity of the catalysts decreased, which meant that Mn impregnation could lead to the deterioration of the framework of Cu-SSZ-13.

H₂-TPR experiments were carried out to evaluate the redox ability of the catalysts, and the results are presented in Figure 4. Three peaks can be observed under 500 °C for Cu-SSZ-13. Peak A was assigned to the reduction of Cu(OH)⁺ species in CHA cages; peak B was attributed to the reduction of CuO_x species; and peak C represented the consumption of H₂ by Cu²⁺ species near d6r [33,34]. It is worth mentioning that no peak corresponding to CuO_x species was observed in the XRD patterns, indicating that these CuO_x species were highly dispersed. In the H₂-TPR curve of 5% Mn/Cu-SSZ-13, five peaks can be observed. The reduction of Cu(OH)⁺ species in CHA cages, CuO_x species and Cu²⁺ species near d6r were also included in Peaks A, B and C [16,33,35,36,37]. However, the peaks for the Cu species shifted to a lower temperature, indicating that the redox ability was improved. Compared to Cu-SSZ-13, there was an increase in the areas of peak A and B, and a decrease in the area of peak C. The decline of peak C can be attributed to a decrement in the amount of Cu²⁺ species near d6r, because the impregnated Mn ions might occupy some ion-exchange sites. Meanwhile, the increase in peak B was partially due to the formation of new CuO species, which emerged because the ion-exchange sites of Cu ions were occupied by Mn ions. During the addition of Mn, Cu(OH)⁺ species in CHA cages must have been lost, since these species are less stable than Cu²⁺ species near d6r. However, the area of peak A increased, which meant that MnO_x species also contributed to peak A. Furthermore, peak D and peak E also emerged as new peaks after the Mn addition, and they were also assigned to the reduction of MnO_x species. In the previous literature, the reduction peaks of pure MnO_x oxides by H₂ were frequently observed in the range between 300 °C and 600 °C, and the reduction of these species occurred by a three-step process: MnO₂→Mn₂O₃→Mn₃O₄→MnO [38,39]. However, with the addition of other elements, the H₂-TPR peaks of MnO_x species could move to much lower temperatures [40,41]. Therefore, in this study, the reduction of MnO₂ to Mn₂O₃ occurred in the range of 150 °C to 350 °C; peak D represented the reduction of Mn₂O₃ to Mn₃O₄, and peak E was assigned to the reduction of Mn₃O₄ to MnO. Impregnation with Mn efficiently improved the redox ability of Cu-SSZ-13. The reaction process of NH₃-SCR over Cu-SSZ-13 was accompanied with an electron transfer between Cu ions and the reactants [9,12,42]. A better redox ability for catalysts might facilitate the electron transfer process, resulting in an improvement in the catalytic activity. As a result, the better redox ability of 5% Mn/Cu-SSZ-13 could be one of the reasons why it exhibited a higher low-temperature activity than Cu-SSZ-13.

XPS was utilized to detect the oxidation state of Cu in Cu-SSZ-13 and Cu & Mn in 5% Mn/Cu-SSZ-13. As presented in Figure S5a, the binding energy values of Cu 2p 1/2 and Cu 2p 3/2 were 953.9 eV and 934.0 eV in both catalysts, respectively. Meanwhile, there were satellite peaks around 965~962 eV and 941~949 eV in 5% Mn/Cu-SSZ-13; however, no satellite peaks were observed in Cu-SSZ-13 [26,27]. Through the deconvolution of the peak with a binding energy of around 940~930 eV, the composition of Cu ions with different valence states was calculated. 75.2% of Cu ions in Cu-SSZ-13 consisted of Cu²⁺, compared with 86.8% Cu²⁺ in 5% Mn/Cu-SSZ-13. A higher amount of Cu²⁺ was beneficial for a low-temperature catalytic activity, so that 5% Mn/Cu-SSZ-13 performed better [26]. Furthermore, the Mn 2p spectra of 5% Mn/Cu-SSZ-13 are presented in Figure S5b, and the binding energy values at 655.3 eV and 643.2 eV were assigned to Mn 2p 1/2 and Mn 2p 3/2, respectively [26,27,39]. To calculate the composition of Mn ions, the peak around 650~640 eV was deconvoluted; the contents of Mn²⁺, Mn³⁺ and Mn⁴⁺ ions were 17.9%, 62.3% and 19.8%, respectively. As a result, in addition to there being more Cu²⁺ ions in 5% Mn/Cu-SSZ-13, another reason why Mn-impregnated catalysts performed better in the low-temperature range was the presence of a large amount of Mn³⁺ and Mn⁴⁺ ions [26,27,39].

To compare the acidity of Cu-SSZ-13 and 5% Mn/Cu-SSZ-13, NH₃-TPD experiments were carried out, and the results are shown in Figure 5. After the deconvolution of the NH₃ desorption profiles, three peaks could be observed for both catalysts. According to the desorption temperature of NH₃, Peak A, Peak B and Peak C were assigned to the desorption of NH₃ from weak acid sites, medium acid sites and strong acid sites, respectively. To further analyze the amount of each kind of acid site, the NH₃ desorption amount per 1 g of catalyst was calculated, and the results are presented in Figure S6. Overall, 5% Mn/Cu-SSZ-13 possessed more acid sites than Cu-SSZ-13, which was attributed to the contribution of the MnO_x species. For Cu-SSZ-13, the weakly adsorbed NH₃ could be attributed to surface hydroxyl species; the medium acid sites were assigned to Lewis acid sites, which were mainly from Cu²⁺ species; and the strong acid sites originated from Brønsted acid sites [43,44,45]. Meanwhile, the desorption of NH₃ from MnO_x species could be observed in the range of 100 °C to 500 °C, so that MnO_x species in 5% Mn/Cu-SSZ-13 contributed to weak, medium and strong acid sites simultaneously [46,47,48]. The amount of weakly absorbed NH₃ that was desorbed was 0.82 mmol/g for 5% Mn/Cu-SSZ-13, while it was 0.55 mmol/g for Cu-SSZ-13. Meanwhile, the amount of NH₃ that was desorbed from strong acid sites was 0.59 mmol/g and 0.44 mmol/g for 5% Mn/Cu-SSZ-13 and Cu-SSZ-13, respectively. However, it is worth mentioning that Cu-SSZ-13 possessed more medium acid sites than 5% Mn/Cu-SSZ-13. This could be due to Cu-SSZ-13 possessing more Cu²⁺ species than 5% Mn/Cu-SSZ-13, which was in accordance with the H₂-TPR results. In conclusion, the increase in the numbers of weak and strong acid sites after the addition of Mn could also be one of the reasons why catalytic activity was promoted.

2.3. Investigation of Reaction Process

To further investigate the NH₃-SCR mechanism over Cu-SSZ-13 and 5% Mn/Cu-SSZ-13, a series of in situ DRIFTS experiments were carried out. The reaction between preabsorbed NH₃ and gaseous NO + O₂ over the catalysts was investigated first, and the results are presented in Figure 6. The peaks around 3400–3200 cm⁻¹ and 1456 cm⁻¹ were assigned to the stretching and bending vibrations of N-H bonds in NH⁴⁺ absorbed on Brønsted acid sites [16,49,50,51]. As the reaction progressed, the peak around 1456 cm⁻¹ first increased and then decreased. The formation of nitrate species resulted in an increase in the peak area around 1456 cm⁻¹, and these species were then consumed at 60 min [49]. The peaks at 3183 cm⁻¹ were assigned to the stretching vibration of NH₃ absorbed on Cu²⁺. Meanwhile, the peaks at 1619 cm⁻¹ and 1255 cm⁻¹ were attributed to the asymmetric and symmetric vibrations of NH₃ absorbed on Lewis acid sites [49,51]. The peak at 1255 cm⁻¹ kept decreasing after the injection of NO and O₂; however, the peak around 1619 cm⁻¹ first decreased and then increased. This phenomenon indicated that the peak around 1619 cm⁻¹ was overlapped by peaks from other species. In other reports, the peaks around 1619 cm⁻¹ were assigned to nitrate species on Cu²⁺ ions and NH₃ simultaneously absorbed on Lewis acid sites [52,53]. Therefore, in this study, the peak around 1619 cm⁻¹ at 60 min was attributed to nitrate species adsorbed on Cu²⁺. Meanwhile, the peak at 1575 cm⁻¹ was also assigned to nitrate species, and nitrate species dominated after NO and O₂ were injected after 60 min [52,53]. The negative peaks at 944 cm⁻¹ and 906 cm⁻¹ were assigned to the T-O-T vibration disturbed by Cu(OH)⁺ and Cu²⁺ species, respectively [16,33,54]. These two peaks’ tendency can be observed for the peak at 1255 cm⁻¹, and almost all NH₃ species absorbed on Cu²⁺ species were consumed at 60 min.

The in situ DRIFTS spectra of 5% Mn/Cu-SSZ-13 are presented in Figure 6b. The characteristic peaks of NH₄⁺ species on the Brønsted acid sites of 5% Mn/Cu-SSZ-13 around 3400–3200 cm⁻¹ and 1457 cm⁻¹ followed the same tendency as observed in Cu-SSZ-13. This indicates that nitrate species are also critical intermediates in the NH₃-SCR process over 5% Mn/Cu-SSZ-13. Apart from Cu²⁺ species, Mnⁿ⁺ species also acted as Lewis acid sites in 5% Mn/Cu-SSZ-13. Unlike the spectra of Cu-SSZ-13, two separate peaks at 1636 cm⁻¹ and 1605 cm⁻¹ assigned to the NH₃ absorbed on Cu²⁺ and Mn ions were observed for 5% Mn/Cu-SSZ-13, respectively [25,26,39,49,51]. The presence of Mn ions influenced the Cu²⁺ species; therefore, the peak for NH₃ absorbed on Cu²⁺ moved to higher wavenumbers when compared with the spectrum of Cu-SSZ-13. Meanwhile, the intensity of negative peaks related to Cu species (around 951 cm⁻¹ and 906 cm⁻¹) decreased when compared with Cu-SSZ-13, which also indicated that Cu²⁺ species were influenced by the addition of Mn. Furthermore, the peak at 1243 cm⁻¹ was assigned to the symmetric vibration of NH₃ absorbed on Lewis acid sites, as indicated in Figure 6a [51]. As the reaction progressed, there was a decrease in the peak areas at 1243 cm⁻¹, 951 cm⁻¹ and 906 cm⁻¹, indicating that NH₃ absorbed on Lewis acid sites was continuously consumed after NO and O₂ were introduced. However, the peak area at 1636 cm⁻¹ increased with the passage of time. This indicated that the peak corresponding to nitrate species overlapped the peak for NH₃ absorbed on Lewis acid sites at 1636 cm⁻¹, and abundant nitrate species accumulated after NO and O₂ were added for 60 min.

To sum up, similar NH₃-SCR mechanisms were observed for Cu-SSZ-13 and 5% Mn/Cu-SSZ-13. With the injection of NO and O₂, nitrate species formed and acted as an important intermediate, indicating that the L-H mechanism took place for both catalysts. Meanwhile, NH₃ absorbed on Lewis acid sites was continuously consumed after NO and O₂ were introduced, which meant that the E-R mechanism also played a role. As a result, the L-H and E-R mechanisms coexisted in the NH₃-SCR process over Cu-SSZ-13 and 5% Mn/Cu-SSZ-13.

The reaction between nitrate species and NH₃ was studied. First, catalysts were exposed to NO and O₂ for 1 h until saturated, and NH₃ was introduced from 0 min after purging by N₂ for 1 h. As can be seen in Figure 7a, the peaks between 1650–1500 cm⁻¹ were assigned to Cu-NO₃⁻ species [49,55,56]. After NH₃ was introduced, the intensity of these peaks decreased, indicating that nitrate species were consumed by NH₃, which further proved that the L-H mechanism existed during NH₃-SCR for the catalysts. The peaks around 3400–3200 cm⁻¹ and 1457 cm⁻¹ were attributed to NH₄⁺ species on Brønsted acid sites and began to emerge at 6 min [16,49,50,51]. Meanwhile, the negative peaks at 951 cm⁻¹ and 906 cm⁻¹, which were respectively assigned to Cu(OH)⁺ and Cu²⁺ species, emerged at 4 min [16,33,54]. Furthermore, new peaks at 1622 cm⁻¹ and 1257 cm⁻¹ emerged at 4 min, and they were assigned to the asymmetric and symmetric vibrations of NH₃ absorbed on Lewis acid sites [49,51]. A new peak emerging at 1622 cm⁻¹ overlapped the peak resulting from nitrate species, but the reaction process could be investigated by observing the peak at 1575 cm⁻¹. After around 16 min, the peak at 1575 cm⁻¹ disappeared, indicating that the absorbed nitrate species were consumed. This process was faster than the reaction between absorbed NH₃ and NO + O₂, as presented in Figure 7. This indicated that, compared with the E-R mechanism, the L-H mechanism proceeded more rapidly for the Cu-SSZ-13 used in this study.

A similar process was observed for 5% Mn/Cu-SSZ-13, as presented in Figure 7b. The peaks around 1650–1500 cm⁻¹ were assigned to Cu-NO₃⁻ species [49,55,56]. Compared with Cu-SSZ-13, more Cu-NO₃⁻ species emerged in the spectrum of 5% Mn/Cu-SSZ-13. Meanwhile, the peak at 1574 cm⁻¹ could hardly be seen after 8 min when compared to 16 min in Cu-SSZ-13, which meant that Cu-NO₃⁻ species were consumed more rapidly with 5% Mn/Cu-SSZ-13. Furthermore, the peaks around 1300 cm⁻¹ and 1200 cm⁻¹ were assigned to bidentate nitrate species in 5% Mn/Cu-SSZ-13, and these species can hardly be seen in Cu-SSZ-13 [55,57]. New peaks at 1632 cm⁻¹, 1604 cm⁻¹ and 1246 cm⁻¹ assigned to NH₃ absorbed on Lewis acid sites emerged starting at 2 min [49,51]. Negative peaks at 946 and 906 representing Cu(OH)⁺ and Cu²⁺ species also appeared at 2 min [16,33,54]. Furthermore, peaks around 3400–3200 cm⁻¹ and 1476 cm⁻¹, which represented NH₄⁺ species absorbed on Brønsted acid sites, also emerged at 4 min [16,49,50,51].

In conclusion, after being exposed to NO and O₂, 5% Mn/Cu-SSZ-13 possessed more nitrate species than Cu-SSZ-13, and these species were consumed more rapidly in 5% Mn/Cu-SSZ-13. As we discussed above, the L-H mechanism played an important role in the NH₃-SCR process of our catalysts, and nitrate species were critical intermediates in the L-H mechanism process. The NH₃-SCR process over Cu-SSZ-13 can be regarded as the redox cycle of Cu⁺/Cu²⁺. Cu²⁺ ions were reduced by NO and NH₃, forming N₂ and H₂O, which consisted of the reduction half cycle. In the oxidation half cycle, two Cu⁺ ions migrated into one cage to activate O₂ and NO, forming nitrate species, and this was the rate-determining step of the whole process [9,12,41,54]. However, with the addition of Mn, the interaction between Mn and Cu facilitated the redox ability of catalysts; meanwhile, MnO_x species are also excellent catalysts for NO oxidation [39], so that nitrates formed more easily in 5% Mn/Cu-SSZ-13 and the former rate-determining step was avoided. As a result, 5% Mn/Cu-SSZ-13 exhibited a better NH₃-SCR performance than Cu-SSZ-13.

2.4. The Stability of 5% Mn/Cu-SSZ-13

To investigate the stability of the catalysts, an NH₃-SCR catalytic activity durability test was carried out, and the results are presented in Figure 8. 5% Mn/Cu-SSZ-13 was chosen as the testing sample due to it having performed the best among all catalysts; meanwhile, the temperature was set at 135 °C when a high NO_x conversion could be obtained and it was not over 100%. After being tested for 24 h, about 90% of NO_x conversion could still be obtained, and no obvious decrease of the catalytic activity was observed. Furthermore, an XRD experiment was also conducted on the fresh 5% Mn/Cu-SSZ-13 catalyst, 5% Mn/Cu-SSZ-13 catalyst tested for NH₃-SCR at 135 °C for 24 h and 5% Mn/Cu-SSZ-13 catalyst tested for NH₃-SCR in the whole temperature range. As can be seen in Figure S7, no obvious decrease of the crystallinity was observed in the catalysts that were either tested at 135 °C for 24 h or tested across the whole temperature range. Durability is of great importance in order for NH₃-SCR catalysts to be utilized in diesel vehicles, and the results indicated that the 5% Mn/Cu-SSZ-13 catalyst possessed excellent stability as well as a bright future in practical applications.

3. Materials and Methods

3.1. Catalyst Preparation

Cu-SSZ-13 catalysts with 4.6 wt% Cu and an Si/Al ratio of 4.8 were prepared by an in situ synthesis method described previously [14,15,58]. To obtain Mn/Cu-SSZ-13 catalysts with a varying Mn content, Cu-SSZ-13 was impregnated with manganese acetate (C₄H₆MnO₄·4H₂O) (SINOPHARM, Beijing, China) solutions of different concentrations. First, 5 g of Cu-SSZ-13 catalyst powder was dispersed in 500 mL deionized water and stirred for 1 h. Then, different amounts of manganese acetate with the equivalent Mn weight of 0.15 g, 0.25 g, 0.35 g, 0.5 g and 0.7 g were added into the solution and stirred for 2 h. The obtained solution was evaporated to dryness by a rotary evaporator at 50 °C, and the powder was placed overnight in an oven at 80 °C. Finally, all the Cu-SSZ-13 powders with and without Mn addition were calcined at 600 °C for 6 h with a ramp rate of 10 °C/min. The obtained catalysts were denoted as Cu-SSZ-13, 3% Mn/Cu-SSZ-13, 5% Mn/Cu-SSZ-13, 7% Mn/Cu-SSZ-13, 10% Mn/Cu-SSZ-13 and 14% Mn/Cu-SSZ-13, respectively.

3.2. Catalyst Characterization

SEM (scanning electron microscopy) (HITACHI, Beijing, China) was utilized to observe the morphology of the catalysts. EDX (Energy Dispersive X-Ray Spectroscopy) (HITACHI, Beijing, China) was used to detect the atoms on the surface of the catalysts. XRD (X-ray diffraction) was carried out to analyze the crystal structure of the catalysts, using a Brucker D8 Advance diffractometer (Brucker, Beijing, China) with Cu Kα (λ = 0.15406 nm). The scan range was from 5°–90°, with step size of 0.02°. N₂ adsorption-desorption was carried out to compare the surface areas and pore volumes of the catalysts before and after Mn addition. The surface area of the catalysts was calculated by the multipore BET method, and the micropore volume was calculated by the t-plot method.

H₂-TPR was carried out to evaluate the redox ability of the catalysts, and the experiment was conducted on a chemisorption analyzer (Micromeritics AutoChem 2920, Beijing, China). First, the catalysts were pretreated under 20% O₂/N₂ at 500 °C for 1 h. After cooling down to room temperature, H₂/Ar was introduced. After the baseline was stable, the temperature was raised to 800 °C with a ramp rate of 10 °C/min. XPS (X-ray photoelectron spectroscopy) (Thermo, Beijing, China) was utilized to detect the valance state of atoms in the catalysts. NH₃-TPD (temperature programmed desorption of NH₃) (Thermo, Beijing, China) was conducted to characterize the acidity of the catalysts. 100 mg of catalyst was placed into a quartz tube sealed at both ends with quartz wool. Catalysts were pretreated at 500 °C under 20% O₂/N₂ for 1 h. After cooling to 50 °C, 500 ppm NH₃/N₂ was introduced until the catalysts were saturated with NH₃, after which the catalysts were purged by N₂ until all the unabsorbed NH₃ was removed. Finally, with a total flow of N₂ of 250 mL/min, the catalysts were heated from 50 °C to 800 °C with a ramp rate of 10 °C/min. The composition of the outlet gas was analyzed by a Nicolet IS 10 infrared spectrometer (Thermo, Beijing, China).

In situ DRIFTS (diffuse reflectance infrared Fourier transform spectroscopy) was used to investigate the reaction mechanism over different catalysts, and the experiments were carried out on a Nicolet IS50 spectrometer (Thermo, Beijing, China) with an MCT/A detector. All catalysts were pretreated at 500 °C for 30 min with 20% O₂/N₂ before the reaction. The whole reaction process was conducted at 150 °C with a total gas flow of 100 mL/min, and the concentrations of NO, NH₃ and O₂ were 1000 ppm, 1000 ppm and 5%, respectively. The first series of experiments were carried out to investigate the reaction between adsorbed NH₃ and gaseous NO + O₂. Catalysts were exposed to NH₃ until saturated and then purged by N₂ to eliminate surplus NH₃. Then, NO and O₂ were introduced, and the spectra were collected. In the second series of experiments, the reaction between adsorbed NO + O₂ and gaseous NH₃ was investigated. NO and O₂ were introduced to the catalysts until saturated, after which the unabsorbed gas was purged by N₂. Finally, NH₃ was introduced, and the spectra were recorded.

3.3. Catalytic Tests

NH₃-SCR tests were carried out in a fixed bed reactor (WeiYe, Beijing, China), with the catalysts loaded in quartz tubes sealed by quartz wool. All the catalysts were sieved to a 40–60 mesh before the NH₃-SCR tests. The total gas flow was 250 mL/min with the gas composition: [NO] = [NH₃] = 500 ppm, [O₂] = 5%, N₂ balance, and GHSV = 100,000 h⁻¹. The outlet gas was detected by a Nicolet IS 10 infrared spectrometer, and the NO_x conversion was calculated by the following equation:

$${\mathrm{NO}}_{x}conversion =\left(1-\frac{{\left[{\mathrm{NO}}_x\right]}_{out}}{{\left[{\mathrm{NO}}_x\right]}_{in}}\right)\times 100\%({\mathrm{NO}}_x=\mathrm{NO}+{\mathrm{NO}}_2)$$

4. Conclusions

In this study, it was found that impregnation with an appropriate amount of Mn could efficiently improve the low-temperature catalytic activity of in situ synthesized Cu-SSZ-13. The promotion effect of Mn on Cu-SSZ-13 was thoroughly investigated from the aspects of the structural framework, redox ability, acid sites and reaction mechanism. It was observed that MnO_x species tended to be located in the pores of Cu-SSZ-13 in a highly dispersed form. Meanwhile, the addition of Mn markedly facilitated the redox ability of Cu-SSZ-13. The improvement of the redox ability promoted the formation of nitrate species, which were critical intermediates in the NH₃-SCR process. As a result, the low-temperature catalytic performance was greatly improved in Mn-impregnated Cu-SSZ-13 catalysts.