English

• The abatement of environmentally harmful NO_x emitted from diesel engines remains challenging for mobile source exhaust purification. Among various techniques, selective catalytic reduction with ammonia (NH₃-SCR) is commercially applied to eliminate the NO_x emitted [1-3]. Compared to other types of catalysts, Cu-based small-pore zeolite catalysts performed higher NO_x activity and hydrothermal stability in NH₃-SCR [4, 5]. In particular, Cu/SSZ-13 has been reported to exhibit high NO_x conversions (> 90%) over a wide temperature range [6]. In Cu/SSZ-13 catalysts, two types of active Cu species exist, including Z₂Cu²⁺ (Z stands for a zeolite framework negative charge) and Z[Cu(OH)]⁺. Among the two active sites, Z[Cu(OH)]⁺ own higher NO_x conversion activity because of the higher mobility [7, 8].

  Despite Cu/SSZ-13 as significant NH₃-SCR catalysts exhibiting good performances and broad application prospects in vehicle NO_x exhaust purification, hydrothermal aging treatment is still inevitable for Cu/SSZ-13 catalysts used in the driving process of diesel vehicles. It is well known that in a typical NH₃-SCR after-treatment system, a diesel particulate filter (DPF) system can remove the unburned particulate matter upstream of the SCR system. During the regeneration process of the DPF system, a considerable amount of heat could be transferred to the SCR catalysts with the steam in the feed [9]. For the carrier of the catalysts, hydrothermal aging treatment leads to the dealumination of the framework, which results in the decrease of Brønsted acid sites. Hydrothermal aging treatment for the active species causes the active Cu ions to transform from ion exchange sites to extra-framework, forming aggregated CuO_x and CuAl₂O₄ species, which results in the decrease of active Cu species [10]. Therefore, modifying the surface of Cu/SSZ-13 catalysts to regulate the stability of Cu active sites and Brønsted acid sites toward high SCR performance.

  Nowadays, zeolite-based core-shell catalysts have attracted more attention. Among the various metal oxide shell materials, ceria (CeO₂) has been proposed to exhibit superior performance in environmental catalysts because of its vital function of storing and releasing oxygen based on the valence transformation of Cerium [11]. However, although the construction of CeO₂ core-shell catalysts of NH₃-SCR reaction is beneficial to improving the catalyst's hydrothermal stability at high temperatures, the reason for the remaining activity of the Cu/SSZ-13 catalyst has yet to be explored.

  Herein, Cu/SSZ-13@CeO₂ was successfully fabricated by self-assembly method (Fig. 1a). Then, Cu/SSZ-13 and Cu/SSZ-13@CeO₂ were hydrothermally treated under NH₃-SCR feed gases at 800 ℃ for 16 h. We chose this deactivated condition because (1) the regeneration of PDF causes the downstream SCR catalyst (Cu/SSZ-13) exposure at high temperature (over 650 ℃) [9]; (2) the hydrothermal aging condition for 16 h corresponding to a 135, 000-mile vehicle-aged catalyst [12]. By combining various catalyst characterizations, the results show that forming the chemical layers (Al−O−Ce) between the zeolite and metal oxide plays a vital role in enhancing the stability of the framework and Brønsted acid sites [13, 14]. The detailed preparation procedures, steps of activity, and characterization methods are presented in the Supporting information.

  Figure 1

  

  Figure 1.  (a) Schematic representation of the preparation of Cu/SSZ-13@CeO₂ catalyst. (b) SEM image of Cu/SSZ-13-F. (c) SEM image of Cu/SSZ-13@CeO₂-F. (d–k) TEM and EDS mapping results of the Cu/SSZ-13@CeO₂-F catalysts. (l) Line scanning of Cu/SSZ-13@CeO₂-F. (m) XRD patterns of the catalysts. (n) 325 nm Raman spectra of the catalysts. (o) N₂ adsorption-desorption isotherms of the catalysts.

  DownLoad: Full-Size Img PowerPoint

  Part of the scanning electron microscope (SEM), transmission electron microscope (TEM), and energy dispersive spectrometer (EDS) mapping results of Cu/SSZ-13-F and Cu/SSZ-13@CeO₂-F are shown in Figs. 1b-l. After CeO₂ cladding Cu/SSZ-13, a uniform particle layer can be observed at the side of Cu/SSZ-13. Line scanning analysis and EDS mapping image results indicated that compared with other elements, Ce was distributed on the side of Cu/SSZ-13, confirming the successful synthesis of Cu/SSZ-13@CeO₂ core-shell catalysts. In addition, the morphology of the other catalysts was investigated through TEM, which is shown in Fig. S1 (Supporting information). All catalysts exhibited a typical cubic nanoparticle. No changes occurred in Cu/SSZ-13-H, indicating that hydrothermal aging did not affect the overall appearance of the catalysts. As expected, for Cu/SSZ-13@CeO₂-F, the CeO₂ shell showed good distributions, which implies that the CeO₂ shell was uniform. Notably, larger particles of CeO₂ formed on the Cu/SSZ-13@CeO₂-H, indicating that part of CeO₂ agglomerated during hydrothermal aging.

  The Cu/SSZ-13 and the core-shell catalysts were checked through XRD detection. As shown in Fig. 1m, characteristic peaks of fresh and deactivated catalysts possessed a typical chabazite pattern (PDF No. 52-0784), indicating that the framework structure of SSZ-13 zeolite was mainly kept intact. For Cu/SSZ-13@CeO₂, four well-resolved diffraction peaks are exhibited indexing as (111), (200), (220), and (311) reflections, which are consistent with the CeO₂ (PDF No. 78-0694). In contrast with Cu/SSZ-13, the weakening of CHA diffractions for Cu/SSZ-13@CeO₂ was simply because of a diluting effect of the CeO₂ shell on the surface of Cu/SSZ-13 in the composite [15]. Compared with fresh Cu/SSZ-13 and Cu/SSZ-13@CeO₂, the diffraction peaks related to CuO (PDF No. 80-1268) and CuAl₂O₄ (PDF No. 75-2360) of the deactivated catalysts cannot be observed, which was probable because they were noncrystalline and/or highly defective. In addition, the diffraction peak intensity of CeO₂ increased after hydrothermal aging treatment, indicating that CeO₂ tended to be agglomerated to larger nanoparticles, which may be caused by high-temperature conditions.

  Next, the UV Raman (325 nm) spectrum of the fresh and deactivated catalysts exhibits a series of bands, which can reveal the framework's structural changes after hydrothermal treatment. As shown in Fig. 1n, five bands were observed for Cu/SSZ-13-F and Cu/SSZ-13-H. The bands at 330 cm⁻¹ can be ascribed to the T−O−T (while T standing Si or Al atoms) banding vibration mode of the six-membered ring (6MR) [16]. The bands at 465 cm⁻¹ and 480 cm⁻¹ can be assigned to the T−O−T bending vibration modes of the 4MR in the D6R and CHA cage, respectively [17]. The bands at 800 cm⁻¹ and 1200 cm⁻¹ can be attributed to the T−O−T stretching vibration modes with symmetric and nonsymmetric, respectively [18]. Each band intensity of Cu/SSZ-13-H was lower than Cu/SSZ-13-F, indicating that the framework was seriously damaged after hydrothermal aging. For Cu/SSZ-13@CeO₂-F and Cu/SSZ-13@CeO₂-H, the bands for T−O−T structure cannot be observed clearly, which was caused by the diluting effect of the CeO₂ shell covering the surface of Cu/SSZ-13 on the catalysts. In addition, a new band at 462 cm⁻¹ was observed in the two catalysts, which was assigned to the F_2g mode of the CeO₂ fluorite phase [19].

  To further understand the structure of the core-shell catalysts, the N₂ adsorption-desorption experiments were tested. As shown in Fig. 1o, based on the International Union of Pure and Applied Chemistry (IUPAC), all the samples exhibited a type Ⅰ isotherm, confirming the existence of a microporous structure. In addition, the mesopores (about 30 nm) are detected in Cu/SSZ-13@CeO₂-F and Cu/SSZ-13@CeO₂-H (Fig. S2 in Supporting information), which was because of the stacking pores formed by CeO₂ on the catalyst surface. Further, the structural properties of the supports are shown in Table S1 (Supporting information). After hydrothermal aging, the S_BET of Cu/SSZ-13-F and Cu/SSZ-13@CeO₂-F decreased gradually from 587.46 m²/g to 550.38 m²/g and 458.21 m²/g to 432.31 m²/g, respectively, indicating that hydrothermal aging seriously caused the collapse of the SSZ-13 framework. It also must be mentioned that there was less Cu/SSZ-13 in Cu/SSZ-13@CeO₂ per unit mass, which led to a smaller S_BET of Cu/SSZ-13@CeO₂. For pore volume (V_t), the V_t of Cu/SSZ-13@CeO₂ increased significantly due to new mesopore formation. For pore diameter (D), Cu/SSZ-13@CeO₂-H exhibited the largest pore diameter (10.22 nm). It indicated that CeO₂ agglomerated to larger nanoparticles, which was consistent with XRD results.

  The NH₃-SCR performance and N₂ selectivity on different catalysts was tested in the temperature range of 100–550 ℃. As shown in Fig. 2a, Cu/SSZ-13-F exhibited 100% NO_x conversion at 250–400 ℃. Compared with Cu/SSZ-13-F, the NO_x conversion of Cu/SSZ-13@CeO₂-F slightly declines at low temperatures (150–250 ℃), which was probable because of the surface blocking after the CeO₂ shell formed on the Cu/SSZ-13. In addition, Fig. S3 (Supporting information) shows the NH₃-SCR performance for CeO₂ under the same GHSV. The results indicated that CeO₂ showed almost no NH₃-SCR activity throughout the test temperature range. After hydrothermal aging, the NO_x conversion of Cu/SSZ-13 (Cu/SSZ-13-H) decreased significantly, almost over 25%. It can be attributed to the loss of acid sites from the zeolite framework and the agglomeration of active Cu sites in the catalyst. Notably, despite the active loss compared with the fresh catalysts, the NO_x conversion of Cu/SSZ-13@CeO₂-H decreased slowly with an active temperature window of 250–450 ℃. It indicated that the CeO₂ shell well protected Cu/SSZ-13 from hydrothermal treatment. Besides, all catalysts exhibit over 90% N₂ selectivity in the whole temperature range, suggesting that the CeO₂ shell had little influence on the N₂ selectivity. In addition, to further investigate the reaction mechanism of the core-shell structure, CeO₂ and CeO₂ + Cu/SSZ-13-H were prepared, and the NH₃-SCR performance and N₂ selectivity on these two deactivated catalysts are shown in Fig. S4 (Supporting information). Compared with Cu/SSZ-13@CeO₂-H, the NO_x conversion of CeO₂ + Cu/SSZ-13-H decreased more significantly indicating that a new chemical bond between CeO₂ and Cu/SSZ-13 may formed in Cu/SSZ-13@CeO₂. Overall, constructing the CeO₂ shell can be an optimal method for the Cu/SSZ-13 catalysts with both well NH₃-SCR performance and higher hydrothermal stability.

  Figure 2

  

  Figure 2.  (a) NO_x conversion and N₂ selectivity as a function of reaction temperature over fresh and deactivated catalysts. (b) Arrhenius plots for standard NH₃-SCR on fresh and deactivated catalysts. The TOF values were calculated by using low NO_x conversions (< 15%) data.

  DownLoad: Full-Size Img PowerPoint

  Fig. 2b shows the catalysts' Arrhenius plots. To avoid the effects of the mass transfer, the NO conversion was adjusted to less than 15% (GHSV = 400, 000 h⁻¹). All catalysts showed almost similar E_a values indicating that the active centers and reaction mechanisms were not changed by the CeO₂ shell on the one hand, and on the other hand by the hydrothermal aging treatment. The preexponential factor of Cu/SSZ-13@CeO₂-F was similar to Cu/SSZ-13-F, indicating that the effectiveness of active sites did not decrease obviously after CeO₂ shell forming. In addition, the preexponential factor of Cu/SSZ-13@CeO₂-H was higher than Cu/SSZ-13-H, suggesting that the CeO₂ shell preserved more effective active sites during the hydrothermal aging.

  To investigate the effect of the zeolite framework after hydrothermal aging, ²⁹Si solid-state NMR was tested. As shown in Fig. 3a, only two features were observed at −108 ppm and −115 ppm, which can be assigned to Si(3Si, 1Al) and Si(4Si, 0Al), respectively [20]. No new peak occurred for Cu/SSZ-13@CeO₂-F, indicating that the introduced CeO₂ did not interact significantly interaction with Si in the framework. Note that the −105 ppm peak of Cu/SSZ-13-H almost disappeared and the −115 ppm peak increased significantly, indicating the occurrence of dealumination in the catalysts, which caused the framework configurations to transform. In contrast, compared with Cu/SSZ-13@CeO₂-F, the change of the two peaks for Cu/SSZ-13@CeO₂-H was inapparent, suggesting the CeO₂ shell formed a protection to the framework structure, especially for Al sites.

  Figure 3

  

  Figure 3.  Characterizations of framework structures and Cu species of the fresh and deactivated catalysts: (a) ²⁹Si NMR spectra; (b) ²⁷Al NMR spectra; (c) H₂-TPR profiles; (d) DRIFTS spectra after NH₃ saturation at 100 ℃; (e, f) EPR spectra.

  DownLoad: Full-Size Img PowerPoint

  ²⁷Al solid-state NMR was carried out to analyze the local Al environment, which is shown in Fig. 3b. Generally, the peaks at 57 ppm and 0 ppm can be attributed to four coordinated framework Al and extra-framework octahedral Al, respectively [21]. The intensity of tetrahedral Al in the framework decreased after hydrothermal aging for Cu/SSZ-13 and Cu/SSZ-13@CeO₂ catalysts, indicative of dealumination. Compared with the core-shell catalyst, more severe dealumination occurred for Cu/SSZ-13. It should be pointed out that all catalysts did not detect extra-framework Al species around 0 ppm, confirming that no Al moved out of the zeolite framework even if the framework fracture happened, which was due to the strong interaction between Al and Cu ions. Besides, for Cu/SSZ-13-H, the peak of the framework Al shifted to 55 ppm. This may be caused by the variation of the local geometry of tetrahedrally coordinated framework Al atoms during the framework collapse [22].

  The existence of active Cu species was conducive to clarifying the influence of core-shell structure after hydrothermal aging treatment. On this basis, H₂-TPR was tested, and the results are shown in Fig. 3c. For Cu species, the peak at 140 ℃ and 250 ℃ can be assigned to the reduction of Z[Cu(OH)]⁺ and Z₂Cu²⁺, respectively [23]. In addition, the peak at 340 ℃ and 420 ℃ can be attributed to the reduction of CuO_x and CuAl₂O₄ species, respectively [24, 25]. For Ce species, the peak at 520 ℃ was related to the surface-capping oxygen of CeO₂ reduced by H₂ and the peaks over 800 ℃ were ascribed to the reduction of bulk CeO₂ [26, 27]. It was clearly observed that more Z[Cu(OH)]⁺ species disappeared in Cu/SSZ-13-H, with the increase of Z₂Cu²⁺, inferring that Z[Cu(OH)]⁺ converted into Z₂Cu²⁺ after hydrothermal aging. However, this phenomenon occurred less frequently in Cu/SSZ-13@CeO₂. The presence of CuO_x and CuAl₂O₄ further demonstrated that hydrothermal aging caused remarkable dealumination and the aggregation of active Cu [10]. Besides, more CuO_x and CuAl₂O₄ clusters forming from Z[Cu(OH)]⁺ and Z₂Cu²⁺ were observed in Cu/SSZ-13-H compared with Cu/SSZ-13@CeO₂-H, which corresponded with the amount changing of Cu ions. Moreover, the bulk CeO₂ was only formed in Cu/SSZ-13@CeO₂-H, indicating that the uniform CeO₂ shell tended to move and form larger particles after hydrothermal aging.

  NH₃-DRIFTS experiments were conducted by exposing the catalysts to NH₃ at 100 ℃ until stabilization. As shown in Fig. 3d, NH₃ adsorption caused the development of several negative going IR bands. These bands can be perturbed by the vibrations arising from Cu species in the close vicinity of the zeolite framework (T−O−T vibrations) [28, 29]. Among the vibrations, the bands at 950 cm⁻¹ and 900 cm⁻¹ were induced by Z[Cu(OH)]⁺ and Z₂Cu²⁺, respectively [23]. For Cu/SSZ-13, after hydrothermal aging, the band of Z[Cu(OH)]⁺ decreased significantly with the increase in band intensity of Z₂Cu²⁺. The result of the transformation between the two Cu species suggested that Z[Cu(OH)]⁺ was more unstable than Z₂Cu²⁺ in Cu/SSZ-13, which was closely consistent with H₂-TPR results. A similar situation also occurred in Cu/SSZ-13@CeO₂. However, the less band intensity loss of Z[Cu(OH)]⁺ for the deactivated catalyst meant that the CeO₂ shell protected the Z[Cu(OH)]⁺ species. In addition, the weakening of the band intensity for Cu/SSZ-13@CeO₂ was primarily due to a diluting effect of the shell phase.

  Further, EPR experiments were conducted to explore the protective effect of CeO₂ shell structure on active Cu species. The double integration of EPR results is proportional to the amount of EPR-active Cu species [30, 31]. The EPR spectra of Cu/SSZ-13 and Cu/SSZ-13@CeO₂ before and after hydrothermal aging treatment were compared separately, which are shown in Figs. 3e and f, respectively. In this case, a single spectral feature was found in the high field, while hyperfine structures were resolved in the low field. After hydrothermal aging, the EPR signal of Cu/SSZ-13 decreased significantly. Assuming Cu/SSZ-13-F contains 100% EPR-active Cu species, the signal loss of Cu/SSZ-13-H was 39.6%. Similarly, compared with Cu/SSZ-13@CeO₂-F, the signal loss of Cu/SSZ-13@CeO₂-H was only 16.3%. It indicated that fewer EPR-active Cu species were lost in Cu/SSZ-13@CeO₂ because of the protection of the core-shell structure, which was consistent with NH₃-SCR performance.

  The formation of a new chemical structure between Ce and zeolite framework presumably caused the reason for Cu/SSZ-13@CeO₂ resisting hydrothermal aging. Therefore, determining the chemical state of Ce helped investigate the reasons for the anti-hydrothermal aging of the catalysts. XPS measurements of Ce element were carried out, and the results are shown in Fig. 4a. The Ce 3d spectra showed a doublet Ce 3d_5/2 and Ce 3d_3/2 separated by ~18.5 eV. Then, the peaks were deconvoluted into eight components corresponding to Ce³⁺ and Ce⁴⁺, respectively [32]. By calculating the deconvoluted peak areas, the amount of Ce³⁺ and Ce⁴⁺ were semi-quantitatively estimated, and the results are shown in Fig. 4b. The percentages of Ce³⁺ and Ce⁴⁺ of CeO₂ were similar to those of Cu/SSZ-13 + CeO₂, indicating that the ratio of Ce³⁺ in CeO₂ mixed with Cu/SSZ-13 by ball milling was not consumed. Compared with Cu/SSZ-13 + CeO₂ and CeO₂, the percentage of Ce³⁺ decreased in both Cu/SSZ-13@CeO₂-F and Cu/SSZ-13@CeO₂-H, confirming that the loss of Ce³⁺ in Cu/SSZ-13@CeO₂ was related to the interaction between the core and shell phase. Additionally, the complex "Ce−O−Z" (Z stands for the carrier of CeO₂) structure had been proven to exist in the surface layers of CeO₂ and other oxide systems containing cerium [33], which further attracted that the Ce³⁺ close to the oxygen vacancies in CeO₂ was consumed by zeolite framework. This reflected that a new structure may form between the CeO₂ and zeolites framework, suppressing the dealumination and protecting the Brønsted acid sites during the hydrothermal aging process.

  Figure 4

  

  Figure 4.  Characterizations of the interaction between Ce species and zeolite SSZ-13 in the fresh and deactivated catalysts: (a) Ce 3d XPS spectra, (b) the percentages of deconvoluted peaks in Fig. 4a, (c) FT-IR spectra, (d) 532 nm Raman spectra. NH₃-DRIFTS spectra results on (e) Cu/SSZ-13-F, (f) Cu/SSZ-13-H, (g) Cu/SSZ-13@CeO₂-F, and (h) Cu/SSZ-13@CeO₂-H. (i) Scheme of the core-shell forming reaction.

  DownLoad: Full-Size Img PowerPoint

  Transmission infrared spectroscopy was measured to understand the effect of the CeO₂ shell on the catalytic activities, and the results are shown in Fig. 4c. The bands associated with the presence of OH- groups in the zeolite structure can also be detected in both Cu/SSZ-13 and Cu/SSZ-13@CeO₂. Notably, compared with the results of Cu/SSZ-13 + CeO₂ and CeO₂ (Fig. S5a in Supporting information), the band at 1400 cm⁻¹ was only found in Cu/SSZ-13@CeO₂-F and Cu/SSZ-13@CeO₂-H. It indicated that a new chemical bond formed in Cu/SSZ-13@CeO₂, which was preliminarily considered to be the vibration of the M-O-Al bridge (M stands for the metal of the shell phase) [34].

  To further investigate the existence of the M-O-Al structure, Raman spectra (532 nm) were tested with the results shown in Fig. 4d. The band at 462 cm⁻¹ was attributed to the F_2g Raman active mode of CeO₂ [35]. The bands at 282 cm⁻¹ and 616 cm⁻¹ were assigned to the A_g and B_g Raman active mode of CuO species, respectively [36]. The band at 566 cm⁻¹ can be ascribed to the stretching vibrations of the M−O−Al bond [13]. For Cu/SSZ-13-H, CuO_x species were detected, suggesting that the active Cu species transformed out from ion exchange sites. Note that a small band appeared in Cu/SSZ-13@CeO₂ at 566 cm⁻¹, which was not observed in the spectra of Cu/SSZ-13 + CeO₂ and CeO₂ (Fig. S5b in Supporting information). This further confirmed the formation of the Ce−O−Al bond between CeO₂ and Al−OH existing in the zeolite framework.

  Next, NH₃-DRIFTS were tested at different temperatures to investigate the details of the reaction. The spectra and corresponding mapping results in the range of 100–500 ℃ are shown in Figs. 4e–h and Fig. S6 (Supporting information). The bands at 1618 cm⁻¹ and 1256 cm⁻¹ can be assigned to NH₃ coordinately linked to Lewis Cu sites and NH₃ wagging mode in Lewis acid sites (labeled as "L-NH₃"), respectively [37]. The bands at 1469, 1200, and 1180 cm⁻¹ can be ascribed to NH₄⁺ species adsorbed on Brønsted acid sites (labeled as "B-NH₃") [38]. For Cu/SSZ-13-H, the B-NH₃ and L-NH₃ bands' intensity decreased significantly, indicating that the acid sites in Cu/SSZ-13 were seriously damaged after hydrothermal aging. Besides, the visible intensity loss of the L-NH₃ adsorption band occurred above 300 ℃, confirming that compared to Cu/SSZ-13-F, the stability of both B-NH₃ and L-NH₃ decreased. For Cu/SSZ-13@CeO₂-F, the bands' intensity of acid sites was not lost obviously. Note that the intensity of B-NH₃ was significantly enhanced, demonstrating that new Brønsted acid sites were formed in the catalysts. Combined with the Ce−O−Al bond observed in Cu/SSZ-13@CeO₂, the reaction occurred by the formation of CeO₂ shell structure (Fig. 4i). The Ce−OH provided new Brønsted acid sites which can promote the enhancement of NO_x conversion and enhance the stability of Al sites in the zeolite framework. In addition, the band at 1480 cm⁻¹ related to monodentate nitrate, which was formed by CeO₂ and decomposed above 200 ℃. For Cu/SSZ-13@CeO₂-H, more bands of monodentate nitrate (1480, 1350, and 1310 cm⁻¹) were observed, which was caused by the agglomeration of CeO₂ after hydrothermal aging. The intensity of L-NH₃ (1618 cm⁻¹) and B-NH₃ (1469 cm⁻¹) bands of Cu/SSZ-13@CeO₂-H was stronger than Cu/SSZ-13-H, indicating that more acidic sites were retained in the catalyst. Besides, NH₃-TPD profiles (Fig. S7 in Supporting information) also showed that compared with Cu/SSZ-13-H, all peaks of Cu/SSZ-13@CeO₂-H exhibited higher signals. It further indicated that part of Cu species and Brønsted acid sites were protected by CeO₂ shell, which was well consistent with NH₃-DRIFTS results.

  The results of NO + O₂-DRIFTS and the corresponding mapping are shown in Fig. S8 (Supporting information). The bands at 1624, 1605/1572, and 1510 cm⁻¹ can be assigned to bridging, bidentate, and monodentate nitrates, respectively [10]. After the forming of the CeO₂ shell, new bands at 1598, 1535/1505 cm⁻¹ were detected, owing to the formation of weakly adsorbed NO₂ on CeO₂ oxide and monodentate nitrates related to Ce species [39, 40]. For Cu/SSZ-13-F, the stabilities of nitrates on active Cu species were in the order of bridging nitrate > bidentate nitrate > monodentate nitrate. The intensity and stability of all nitrate bands decreased significantly after hydrothermal aging treatment, confirming the loss of active Cu species. Compared with Cu/SSZ-13-F, more bidentate and monodentate nitrates with higher stability were formed in Cu/SSZ-13@CeO₂-F, indicating that more acid sites were provided by the CeO₂ shell structure. Different from Cu/SSZ-13-H, Cu/SSZ-13@CeO₂-H showed higher nitrate bands signal as well as better stability. This indicated that the CeO₂ shell structure protected active Cu species from hydrothermal aging and provided more acid sites, which enhanced the anti-hydrothermal aging of Cu/SSZ-13.

  Fig. S9 (Supporting information) exhibited in situ DRIFTS spectra recorded at 175 ℃ under the treatment that NH₃ first treated the catalysts, then NO + O₂ was added. Among all catalysts, the bands of L-NH₃ (1618 cm⁻¹ and 1256 cm⁻¹) and B-NH₃ (1469 cm⁻¹ and 1180 cm⁻¹) were detected at 0 min, similar to NH₃-DRIFTS results. After introducing NO+O₂, the bands of acid sites decreased with the formation of nitrates, confirming the reaction between adsorbed NH₃ and NO_x. Furthermore, the gases were introduced reversely in the system, and the results are shown in Fig. S10 (Supporting information). The nitrates formed in all catalysts at 0 min and gradually disappeared after NH₃ was introduced, with the appearance of bands for acid sites. This reflected that the NH₃-SCR reaction pathways of all catalysts follow the Langmuir–Hinshelwood (L–H) mechanism, which was consistent with the kinetic analysis results.

  In conclusion, this work successfully demonstrated that the hydrothermal stability of Cu/SSZ-13 catalysts for selective catalytic reduction (SCR) in diesel exhaust systems can be significantly enhanced through the application of a novel CeO₂ core-shell structure. As shown in Fig. 5, the Cu/SSZ-13@CeO₂ composite material exhibits superior resistance to hydrothermal aging, maintaining functionality and preventing the deactivation of [Cu(OH)]⁺ active sites. Additionally, the CeO₂ shell effectively preserves both Lewis and Brønsted acid sites, which are essential for sustained SCR performance. A pivotal advancement is the formation of a "Ce−O−Al" bond between the CeO₂ shell and the Cu/SSZ-13 core. This chemical bond crucially enhances the structural integrity of the zeolite framework, thereby augmenting the catalyst's durability under harsh operational conditions typical of diesel vehicle exhaust systems. These findings not only advance our understanding of material science in catalysis but also pave the way for the development of more robust SCR catalysts, capable of meeting the stringent environmental regulations for diesel emissions.

  Figure 5

  

  Figure 5.  Mechanism of the enhancement of hydrothermal stability over the Cu/SSZ-13@CeO₂ catalysts.

  DownLoad: Full-Size Img PowerPoint

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Jianning Zhang: Writing – original draft, Methodology, Investigation, Conceptualization. Yihuai Zhang: Investigation. Guoxin Ma: Investigation. Jingchen Zhao: Investigation. Tao Zhang: Writing – review & editing, Visualization, Resources, Methodology, Funding acquisition, Conceptualization. Jian Liu: Supervision, Funding acquisition.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (No. 22176216), the National Engineering Laboratory for Mobile Source Emission Control Technology (No. NELMS2020A13), and the National Key Research and Development Program of China (No. SQ2022YFB3500058).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110516.

  
  
• 1. [1]

     L. Han, S. Cai, M. Gao, et al., Chem. Rev. 119 (2019) 10916–10976. doi: 10.1021/acs.chemrev.9b00202

  2. [2]

     M.V. Twigg, Appl. Catal. B 70 (2007) 2–15. doi: 10.1016/j.apcatb.2006.02.029

  3. [3]

     R. Zhang, N. Liu, Z. Lei, et al., Chem. Rev. 116 (2016) 3658–3721. doi: 10.1021/acs.chemrev.5b00474

  4. [4]

     L. Ma, Y. Cheng, G. Cavataio, et al., Appl. Catal. B 156–157 (2014) 428–437. doi: 10.1016/j.apcatb.2014.03.048

  5. [5]

     Y.J. Kim, J.K. Lee, K.M. Min, et al., J. Catal. 311 (2014) 447–457. doi: 10.1016/j.jcat.2013.12.012

  6. [6]

     M. Jabłońska, RSC Adv. 12 (2022) 25240–25261. doi: 10.1039/d2ra04301g

  7. [7]

     K.A. Lomachenko, E. Borfecchia, C. Negri, et al., J. Am. Chem. Soc. 138 (2016) 12025–12028. doi: 10.1021/jacs.6b06809

  8. [8]

     F. Gao, D. Mei, Y. Wang, et al., J. Am. Chem. Soc. 139 (2017) 4935–4942. doi: 10.1021/jacs.7b01128

  9. [9]

     D. Wang, Y. Jangjou, Y. Liu, et al., Appl. Catal. B 165 (2015) 438–445. doi: 10.1016/j.apcatb.2014.10.020

  10. [10]

      Y. Zhang, H. Zhu, T. Zhang, et al., Environ. Sci. Technol. 56 (2022) 1917–1926. doi: 10.1021/acs.est.1c06068

  11. [11]

      T. Xu, G. Li, X. Zhu, et al., Fuel 353 (2023) 129292. doi: 10.1016/j.fuel.2023.129292

  12. [12]

      S.J. Schmieg, S.H. Oh, C.H. Kim, et al., Catal. Today 184 (2012) 252–261. doi: 10.1016/j.cattod.2011.10.034

  13. [13]

      L. Wu, B. Peng, Q. Li, et al., Environ. Sci. Pollut. R. 26 (2019) 19665–19675. doi: 10.1007/s11356-019-05161-7

  14. [14]

      A. Mitra, M.K. Mishra, J. Saha, Micropor. Mesopor. Mat. 203 (2015) 151–162. doi: 10.1016/j.micromeso.2014.10.014

  15. [15]

      T. Zhang, F. Qiu, J. Li, Appl. Catal. B 195 (2016) 48–58. doi: 10.1016/j.apcatb.2016.04.058

  16. [16]

      P.K. Dutta, D.C. Shieh, M. Puri, Zeolites 8 (1988) 306–309. doi: 10.1016/S0144-2449(88)80127-1

  17. [17]

      J. Zhang, Y. Chu, X. Liu, et al., Chin. J. Catal. 40 (2019) 1854–1859. doi: 10.1016/S1872-2067(19)63287-0

  18. [18]

      P.K. Dutta, M. Puri, J. Phys. Chem. A 91 (1987) 4329–4333. doi: 10.1021/j100300a025

  19. [19]

      Z. Wu, M. Li, J. Howe, et al., Langmuir 26 (2010) 16595–16606. doi: 10.1021/la101723w

  20. [20]

      K.U. Gore, A. Abraham, S.G. Hegde, et al., J. Phys. Chem. B 106 (2002) 6115–6120. doi: 10.1021/jp0143241

  21. [21]

      J. Song, Y. Wang, E.D. Walter, et al., ACS Catal. 7 (2017) 8214–8227. doi: 10.1021/acscatal.7b03020

  22. [22]

      R.A. Beyerlein, C. Choi-Feng, J.B. Hall, et al., Top. Catal. 4 (1997) 27–42. doi: 10.1023/A:1019188105794

  23. [23]

      J.H. Kwak, H. Zhu, J.H. Lee, et al., Chem. Commun. 48 (2012) 4758–4760. doi: 10.1039/c2cc31184d

  24. [24]

      B. Mark, H.P. Jacqueline, A.S. Valerie, et al., Appl. Catal. 58 (1990) 147–153. doi: 10.1016/S0166-9834(00)82284-8

  25. [25]

      T. Sutarat, K. Sirapassorn, J. Anchalee, et al., Catal. Commun. 169 (2022) 106468. doi: 10.1016/j.catcom.2022.106468

  26. [26]

      H. Zhu, Z. Qin, W. Shan, et al., J. Catal. 225 (2004) 267–277. doi: 10.1016/j.jcat.2004.04.006

  27. [27]

      W. Shan, H. Guo, C. Liu, et al., J. Rare Earth 30 (2012) 665–669. doi: 10.1016/S1002-0721(12)60109-4

  28. [28]

      J.H. Kwak, T. Varga, C.H.F. Peden, et al., J. Catal. 314 (2014) 83–93. doi: 10.1016/j.jcat.2014.03.003

  29. [29]

      J. Luo, D. Wang, A. Kumar, Catal. Today 267 (2016) 3–9. doi: 10.1016/j.cattod.2015.12.002

  30. [30]

      A. Godiksen, F.N. Stappen, P.N.R. Vennestrøm, et al., J. Phys. Chem. C 118 (2014) 23126–23138. doi: 10.1021/jp5065616

  31. [31]

      Y. Zhang, Y. Wu, Y. Peng, et al., J. Phys. Chem. C 124 (2020) 28061–28073. doi: 10.1021/acs.jpcc.0c07971

  32. [32]

      J.P. Holgado, R. Alvarez, G. Munuera, Appl. Surf. Sci. 161 (2000) 301–315. doi: 10.1016/S0169-4332(99)00577-2

  33. [33]

      A.R. González-Elipe, A. Fernández, J.P. Holgado, et al., J. Vac. Sci. Technol. A 11 (1993) 58–65. doi: 10.1116/1.578720

  34. [34]

      M. Mazloum-Ardakani, F. Sabaghian, M. Yavari, et al., J. Alloys Compd. 819 (2020) 152949. doi: 10.1016/j.jallcom.2019.152949

  35. [35]

      B.M. Reddy, A. Khan, P. Lakshmanan, et al., J. Phys. Chem. B 109 (2005) 3355–3363. doi: 10.1021/jp045193h

  36. [36]

      M. Rashad, M. Rüsing, G. Berth, et al., J. Nanomater. 2013 (2013) 82. doi: 10.1155/2013/714853

  37. [37]

      I. Lezcano-Gonzalez, U. Deka, B. Arstad, et al., Phys. Chem. Chem. Phys. 16 (2014) 1639–1650. doi: 10.1039/C3CP54132K

  38. [38]

      C. Negri, E. Borfecchia, M. Cutini, et al., ChemCatChem 11 (2019) 3828–3838. doi: 10.1002/cctc.201900590

  39. [39]

      G. Qi, R.T. Yang, R. Chang, Appl. Catal. B 51 (2004) 93–106. doi: 10.1016/j.apcatb.2004.01.023

  40. [40]

      B. Liu, J. Xin, T. Zhang, et al., ACS Catal. 11 (2021) 7613–7636. doi: 10.1021/acscatal.1c00311

• 

  Figure 1  (a) Schematic representation of the preparation of Cu/SSZ-13@CeO₂ catalyst. (b) SEM image of Cu/SSZ-13-F. (c) SEM image of Cu/SSZ-13@CeO₂-F. (d–k) TEM and EDS mapping results of the Cu/SSZ-13@CeO₂-F catalysts. (l) Line scanning of Cu/SSZ-13@CeO₂-F. (m) XRD patterns of the catalysts. (n) 325 nm Raman spectra of the catalysts. (o) N₂ adsorption-desorption isotherms of the catalysts.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) NO_x conversion and N₂ selectivity as a function of reaction temperature over fresh and deactivated catalysts. (b) Arrhenius plots for standard NH₃-SCR on fresh and deactivated catalysts. The TOF values were calculated by using low NO_x conversions (< 15%) data.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Characterizations of framework structures and Cu species of the fresh and deactivated catalysts: (a) ²⁹Si NMR spectra; (b) ²⁷Al NMR spectra; (c) H₂-TPR profiles; (d) DRIFTS spectra after NH₃ saturation at 100 ℃; (e, f) EPR spectra.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Characterizations of the interaction between Ce species and zeolite SSZ-13 in the fresh and deactivated catalysts: (a) Ce 3d XPS spectra, (b) the percentages of deconvoluted peaks in Fig. 4a, (c) FT-IR spectra, (d) 532 nm Raman spectra. NH₃-DRIFTS spectra results on (e) Cu/SSZ-13-F, (f) Cu/SSZ-13-H, (g) Cu/SSZ-13@CeO₂-F, and (h) Cu/SSZ-13@CeO₂-H. (i) Scheme of the core-shell forming reaction.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Mechanism of the enhancement of hydrothermal stability over the Cu/SSZ-13@CeO₂ catalysts.

  下载: 全尺寸图片 幻灯片
•