Engineering In-Co₃O₄/H-SSZ-39(OA) Catalyst for CH₄-SCR of NO_x: Mild Oxalic Acid (OA) Leaching and Co₃O₄ Modification

Abstract

Zeolite-based catalysts efficiently catalyze the selective catalytic reduction of NO_x with methane (CH₄-SCR) for the environmentally friendly removal of nitrogen oxides, but suffer severe deactivation in high-temperature SO₂- and H₂O-containing flue gas. In this work, SSZ-39 zeolite (AEI topology) with high hydrothermal stability is reported for preparing CH₄-SCR catalysts. Mild acid leaching with oxalic acid (OA) not only modulates the Si/Al ratio of commercial SSZ-39 to a suitable value, but also removes some extra-framework Al atoms, introducing a small number of mesopores into the zeolite that alleviate diffusion limitation. Additional Co₃O₄ modification during indium exchange further enhances the catalytic activity of the resulting In-Co₃O₄/H-SSZ-39(OA). The optimized sample exhibits remarkable performance in CH₄-SCR under a gas hourly space velocity (GHSV) of 24,000 h⁻¹ and in the presence of 5 vol% H₂O. Even under harsh SO₂- and H₂O-containing high-temperature conditions, it shows satisfactory stability. Catalysts containing Co₃O₄ components demonstrate much higher CH₄ conversion. The strong mutual interaction between Co₃O₄ and Brønsted acid sites, confirmed by the temperature-programmed desorption of NO (NO-TPD), enables more stable N_xO_y species to be retained in In-Co₃O₄/H-SSZ-39(OA) to supply further reactions at high temperatures.

1. Introduction

The selective catalytic reduction of NO_x with CH₄ (CH₄-SCR) has attracted considerable interest because it is capable of simultaneously abating harmful NO_x and unburned CH₄ emissions from vehicle and power plant exhaust [1]. Metal-exchanged zeolite catalysts with relatively high catalytic activity over a wide temperature range are extensively studied in the CH₄-SCR reaction. However, they suffer from poor hydrothermal stability, exhibiting a considerable activity decrease in the presence of high-temperature water vapor, especially those that are Al-rich and have large pores [2,3,4,5,6,7,8,9,10]. In the NH₃-SCR field where ammonia is employed as a reducing agent, Cu-exchanged small-pore SSZ-13 zeolites (CHA topology) have been implemented as a new-generation catalyst in diesel after-treatment systems due to their high deNO_x activity and good hydrothermal stability [11]. Recently, an alternative small-pore SSZ-39 zeolite (AEI topology), which has a different connection mode of neighboring double six-ring (d6r), demonstrated even better hydrothermal stability in NH₃-SCR reactions [12]. Superior hydrothermal stability is a crucial requirement for a favorable CH₄-SCR reaction, as it usually occurs at a relatively high temperature and the real exhaust always contains a certain amount of water vapor, exposing the catalyst to hydrothermal conditions during operation [13].

In our previous studies, we synthesized indium-exchanged zeolites for CH₄-SCR and discovered that the framework type [14] and Si/Al ratio [15] of zeolites affected deNO_x performance. The Si/Al ratio of a zeolite remarkably affects its acidity and stability. Zeolites with low Si/Al ratios possess abundant Brønsted acid sites (BASs, formed by protons compensating the negatively charged O atoms induced by the substitution of Si atoms by Al^IV atoms in the framework) serving as ion-exchange/active sites, but are more susceptible to dealumination under high-temperature hydrothermal conditions [12,16,17,18]. Steaming and/or acid leaching is an effective method for the selective extraction of Al atoms from the zeolite framework, enabling the convenient decrease of excessive acid densities and the modulation of Si/Al ratios [19]. Maintaining sufficiently enough exchangeable sites and zeolite integrity, however, needs to be considered in the dealumination treatment. Thus, carefully manipulating the mild dealumination of Al-rich zeolite is highly desirable so as to simultaneously achieve medium acidity and the preferred durability.

In the present work, commercial Al-rich H-SSZ-39 was dealuminated with oxalic acid (OA) and ion-exchanged with an indium nitrate aqueous solution to obtain In/H-SSZ-39(OA), exhibiting excellent deNO_x activity in a dry CH₄-SCR reaction. Although the Pristine SSZ-39 showed negligible deNO_x activity after In exchange, the introduction of mild dealumination via weak acid etching prior to In exchange greatly improved the catalytic performance in the CH₄-SCR reaction. Moreover, introducing a small amount of Co₃O₄ fine powder into the indium nitrate solution further enhanced the deNO_x activity of the resulting In-Co₃O₄/H-SSZ-39(OA) catalysts. The efficacy of mild acid etching post-treatment and Co₃O₄ modification was tentatively illustrated based on comprehensive catalyst structure and reaction pathway analysis. ²⁷Al and ²⁹Si NMR, XRD, and N₂ adsorption–desorption were conducted to characterize the zeolite framework structure changes. Microscopy, XPS, NH₃-TPD, and NO-TPD were used to investigate the In/Co distributions, surface acid sites, and active intermediate N_xO_y species.

2. Results and Discussion

2.1. Catalytic Activity

2.1.1. CH₄-SCR Activity of In/H-SSZ-39(OA) Catalysts

The catalytic activity of In-exchanged Pristine H-SSZ-39 (In/H-SSZ-39) and a series of In/H-SSZ-39(OA) catalysts in CH₄-SCR reaction was investigated. In-free Pristine H-SSZ-39 and H-SSZ-39(OA) samples were also tested for CH₄-SCR reaction; however, they showed quite low deNO_x activity (Figure S1a), indicating that the introduced indium species acted as the active component of the catalysts. As shown in Figure 1a, In/H-SSZ-39 without acid pretreatment exhibited very limited activity with <15% NO_x conversion at 550 °C, while In/H-SSZ-39(OA) catalysts demonstrated considerably enhanced activity with >40% NO_x conversion at 550 °C. Among them, the catalyst sample made from H-SSZ-39 etched with 1.0 M OA solution showed the highest NO_x conversion (∼60% at 550 °C). Elevating the reaction temperature above 550 °C declined NO_x conversion due to the non-selective oxidation of CH₄, evidenced by the significantly enhanced CH₄ conversion above 550 °C in Figure 1d–f and the dramatic difference between the CH₄ selectivities at 550 °C and 600 °C in Figure 1g–i. Too low or high a concentration of OA led to an insignificant or excessive etching effect, detrimental to the CH₄-SCR activity. Thus, the OA concentration in the SSZ-39 dealumination step was fixed at 1.0 M, and the effects of etching time and etching temperature were further investigated. As for the etching time, 4 h was found to be optimal (Figure 1b,e). The etching time of 2 h only slightly enhanced the deNO_x activity of the catalyst. Extending the etching time to more than 4 h led to a gradual decrease in catalytic activity, and at the extreme of 10 h, the extensive etching yielded a low-efficiency catalyst that showed even more inferior performance than the In/H-SSZ-39 sample without etching pretreatment. A reasonable explanation could be that as etching proceeded, massive dealumination resulted in reduced exchangeable sites for In species in SSZ-39 zeolite and collapsed the zeolite framework. Compared to the significant effect of etching time, the etching temperature only have a minor effect on the catalytic activity, with the sample etched at 80 °C being slightly better than the others (Figure 1c,f).

2.1.2. CH₄-SCR Activity of In-Co₃O₄/H-SSZ-39(OA) Catalysts

Previous studies have demonstrated that introducing small amounts of metal oxides (especially, Co₃O₄) into In-exchanged zeolite catalytic systems can improve CH₄-SCR activity under harsh SO₂- and H₂O-containing conditions [20,21]. Therefore, small amounts of Co₃O₄ were dispersed during the In-exchange process on optimally acid-etched H-SSZ-39, with pretreatment using 1.0 M of OA at 80 °C for 4 h. The Co₃O₄-modifying amount for preparing In/H-SSZ-39(OA) catalysts (denoted as In-Co₃O₄/H-SSZ-39(OA)) in wet CH₄-SCR reaction was investigated, as shown in the comparative catalytic activities (Figure 2a,b). Modification with trace amounts of Co₃O₄ (Co₃O₄ to zeolite mass ratio between 1/30–1/50) significantly boosted the highest NO_x conversion to >70% and facilitated CH₄ conversion even in the presence of 5 vol% H₂O. The In-free sample of Co₃O₄/H-SSZ-39(OA) demonstrated boosted CH₄ conversion compared to H-SSZ-39(OA) (Figure S1b), further proving that Co₃O₄ could promote CH₄ conversion. Moreover, its negligible CH₄-SCR activity (Figure S1a) signified that Co₃O₄ did not serve as a second active center responsible for NO_x reduction but rather acted as a promoter. The highest NO_x conversion over In/H-SSZ-39(OA) occurred at higher temperatures (∼600 °C), and the optimal amount of Co₃O₄ was determined to be 30:1, for this sample exhibited the highest NO_x conversion of ∼83% (Figure 2a). The Co₃O₄-modified catalysts demonstrated higher NO_x and CH₄ conversion compared to Co₃O₄-free sample. Elevating the Co₃O₄ amount with a mass_zeolite: ${\mathrm{mass}}_{\begin{array}{c}\hfill errortypeceCo3O4\end{array}}$ ratio from 1:50 to 1:30 greatly enhanced NO_x and CH₄ conversions. A further increase in Co₃O₄ amount obviously lowered the NO_x conversion. The CH₄ conversions over In-Co₃O₄/H-SSZ-39(OA) catalysts under wet conditions were even higher than those of Co₃O₄-free ones under dry conditions (Figure 1d–f). The sample modified by the largest amount of Co₃O₄ (5:1) exhibited the highest CH₄ conversion at 400–650 °C, but demonstrated the worst NO_x conversion at a temperature range of 550–650 °C in CH₄-SCR reaction, which was presumably related to the non-selective oxidation of methane catalyzed by excessive Co₃O₄ (Figure S2).

An In-Co₃O₄/H-SSZ-39(OA) catalyst prepared using the optimized OA etching conditions (0.1 M OA, 80 °C, 4 h) and Co₃O₄ dosage (Co₃O₄: H-SSZ-39(OA) mass ratio of 1:30) was tested under different CH₄-SCR reaction conditions, as shown in Figure S3. Operation parameters including O₂ concentration, CH₄/NO ratio, H₂O concentration, and GHSV all affected NO_x and CH₄ conversions as well as CH₄ selectivity, with the effect of GHSV being most significant. Under a GHSV of 12,000 h⁻¹, the highest NO_x conversion of ∼88% occurred at 600 °C. The deNO_x activity and CH₄ selectivity roughly showed an increasing and then decreasing trend with O₂ concentration, with the turning point occurring at an O₂ concentration of 10 vol%. Higher CH₄/NO ratios resulted in declined CH₄ selectivity but slightly improved deNO_x activity. The high concentration of water vapor adversely affected the catalyst, as evidenced by the continuously decreasing NO_x conversion and CH₄ selectivity with higher water vapor concentrations, while the CH₄ conversion was largely unaffected. The tolerance to SO₂ was also tested, and the In-Co₃O₄/H-SSZ-39(OA) catalyst maintained its high activity when the SO₂ concentration was 50 ppm and 5 vol% water vapor was present (Figure S4). Therefore, even being operated under SO₂- and H₂O-containing conditions, the In-Co₃O₄/H-SSZ-39(OA) catalyst demonstrated excellent recyclability. The maximum NO_x conversion of the catalyst in the third TPSR test was ∼50%, which was only ∼20% lower than that in the first test (Figure 2c). The catalyst also showed high stability, as shown in Figure 2d, the NO_x conversion over the In-Co₃O₄/H-SSZ-39(OA) catalyst showed a slow downward trend for the first four hours and then gradually stabilized afterwards (∼55% at 600 °C). SO₂ poisoning is mostly associated with the formation of sulphate species under oxidizing conditions [22], whereas H₂O vapor usually led to the aggregation and sintering of active sites in the zeolite-based catalysts, forming weakly active or inactive metal oxide clusters [23,24].

2.2. Catalyst Characterization

2.2.1. Microscopy

In the SEM images of Pristine H-SSZ-39 (Figure 3a), cuboid particles with a mean size of ∼1 $\mathsf{\mu }$m could be observed. Pristine SSZ-39 generally appeared as intact crystals with smooth surfaces and distinct edges. After acid etching and In exchange, the resulting In/H-SSZ-39(OA) exhibited mostly integrated crystals but partly with missing edges or surface depressions, suggesting that etching might start from the crystal periphery (Figure 3b). Elemental compositions determined by ICP-OES and elemental distributions of samples at different preparation stages were shown in

Table 1. Physicochemical parameters of Pristine H-SSZ-39, In/H-SSZ-39(OA), and In-Co₃O₄/H-SSZ-39(OA).

Sample	Relative Crystallinity a (%)	SBET b (m2 g−1)	Vtot c (cm3 g−1)	Vmicro d (cm3 g−1)	Vmeso e (cm3 g−1)	Si/Al Bulk f	In Content f (wt%)	Co Content f (wt%)
Pristine H-SSZ-39	100.0	746.7	0.293	0.278	0.015	4.72	/	/
In/H-SSZ-39(OA)	72.0	699.5	0.303	0.242	0.060	6.79	4.40	/
In-Co3O4/H-SSZ-39(OA)	33.4	566.6	0.272	0.202	0.070	6.86	7.17	1.14

^a Calculated from the sum of the integral areas of diffraction peaks ascribed to (111), (200), (113), (310), (132), (133), and (025) crystal planes. ^b Calculated by the Brunauer–Emmett–Teller (BET) model. ^c Calculated from the adsorption amount at a relative pressure (P/P₀) close to 1. ^d Calculated using the t-plot method. ^e Calculated as the difference between V_tot and V_micro. ^f Determined by ICP-OES.

and Figures S5–S10. In species were uniformly distributed in In/H-SSZ-39(OA), without large indium oxide particles being observed (Figures S7 and S9); whereas for In-Co₃O₄/H-SSZ-39(OA), a few large indium and cobalt oxide particles were present (Figure 3c, S8, and S10). In a representative HRTEM image of In/H-SSZ-39(OA) (Figure 3e), some nanoparticles attached to the zeolite surface with a lattice spacing of ∼0.293 nm assigned to the cubic In₂O₃ (c-In₂O₃) (222) crystal plane could be observed. Meanwhile, some mesopores appeared in the In/H-SSZ-39(OA) zeolite (Figure S11b), which contrasted with the Pristine SSZ-39 (Figure 3d and S11a). In-Co₃O₄/H-SSZ-39(OA) contained apparently broken crystal fragments with irregular shapes and a more obvious mesopore structure (Figure 3f and S11c). As shown in Figure 3f, c-In₂O₃, rhombohedral In₂O₃ (rh-In₂O₃), and Co₃O₄ nanoparticles were distributed on In-Co₃O₄/H-SSZ-39(OA), as confirmed by lattice fringes attributed to c-In₂O₃ (222) (d = 0.292 nm), c-In₂O₃ (400) (d = 0.252 nm), rh-In₂O₃ (104) (d = 0.289 nm), and Co₃O₄ (220) (d = 0.285 nm). The zeolite framework, with lattice fringes (d = 0.911 nm) assigned to AEI (110) or (002) crystal planes were the bulk phase, as shown in Figure S11c.

2.2.2. Crystalline Properties

As observed from Figure 3g, PXRD patterns demonstrated that Pristine H-SSZ-39 was a phase-pure AEI zeolite that matched well with the simulated structure. After acid leaching, the diffraction peaks slightly shifted toward higher 2$\theta$ values, suggesting a lattice contraction as a result of a change in the chemical composition, that was an extraction of Al from the unit cell of SSZ-39 (Figure S12a). Accordingly, the relative crystallinity slightly decreased to 92.5% after OA leaching. Indium exchange introduced In atoms that are larger than the host atoms, leading to lattice expansion, and thus the diffraction peaks slightly shifted toward lower 2$\theta$ values. Modification with Co₃O₄ did not result in any shift in the diffraction peaks, signifying that Co₃O₄ might not incorporate into the interior of SSZ-39 zeolite. It had been reported that the migration of In species would have an impact on the zeolite channel structure [25], which might contribute to the decrease in relative crystallinity after In exchange (Table 1). In species (e.g., In₂O₃) were undetectable by XRD for In/H-SSZ-39(OA), likely because of their low loading, small size, and highly dispersed distribution, consistent with the EDS mapping results (Figures S7 and S9). Distinct characteristic diffraction peaks of c-In₂O₃ and rh-In₂O₃ could be observed from the PXRD pattern of In-Co₃O₄/H-SSZ-39(OA) (Figure 3g, S12c,d, and S13b), which might be attributed to the reduced exchangeable sites in the partly amorphized structure of this sample (relative crystallinity = 33.4%, Table 1). In species that exceeded the exchange capacity of zeolites with reduced crystallinity formed extra-zeolite In₂O₃ particles during calcination, which were typically inactive in the catalyzing CH₄-SCR reaction [26,27]. Moreover, diffraction peaks attributed to Co₃O₄ were found for In-Co₃O₄/H-SSZ-39(OA) in Figure S12b, even though some of the diffraction peaks of multiple components overlapped each other and those from the Co₃O₄ component were very weak. No detectable indium oxides and cobalt oxides by XRD other than In₂O₃ and Co₃O₄ were present (Figure S13), which corresponded well with HRTEM observation.

2.2.3. Textural Properties

The samples were characterized by N₂ adsorption–desorption isotherm measurements at 77 K (Figure 3h), and their textural properties including non-local density functional theory (NLDFT) pore size distributions (PSD) are shown in Table 1 and Figure 3i. Pristine H-SSZ-39 exhibited a typical type I isotherm, characterized by a sharp rise in N₂ adsorption at low pressure (p/p₀ < 0.01) followed by a saturation plateau, indicating the predominance of microporosity. No obvious hysteresis loop was observed, corresponding to its negligible mesoporosity (Table 1 and Figure 3i). The condensation of N₂ molecules in interparticle voids contributed to a minor increase in adsorption at p/p₀ ≈ 1.0. Acid etching and In exchange slightly decreased the microporosity while increasing the mesoporosity, indicating that a small number of neighboring micropores merged to form mesopores during Al extraction and In migration. In/H-SSZ-39(OA) had a similar type I isotherm but a more pronounced condensation of N₂ molecules, which might be associated the increased mesoporosity in the zeolite (Figure 3i inset and Table 1). This phenomenon was most noticeable in In-Co₃O₄/H-SSZ-39(OA), and further amorphorized zeolite fragments might also make a contribution, consistent with the microscopy observation and PXRD measurement. In addition, In-Co₃O₄/H-SSZ-39(OA) had the lowest low-pressure N₂ adsorption capacity, corresponding to its smallest micropore volume (0.202 cm³ g⁻¹). As shown in Figure 3i, the mesopores in In/H-SSZ-39(OA) and In-Co₃O₄/H-SSZ-39(OA) were mainly distributed in 20–50 nm. The constructed hierarchical structure might contribute to the enhanced catalytic activity through enhanced mass transfer.

2.2.4. Coordination Environment of T-Atoms

Figure 4a depicted ²⁷Al MAS SSNMR spectra of Pristine H-SSZ-39 and In/H-SSZ-39(OA). For Pristine H-SSZ-39, the sharp signal at ∼60 ppm was associated with tetrahedra framework Al (FAl) sites, designated as Al(IV)-1. In addition, a broad peak centered at ∼52 ppm was attributed to the partially coordinated FAl atoms with hydroxyl groups (${\left(\mathrm{SiO}\right)}_{4-\mathit{n}}$−Al(OH)$\mathit{n}$, n = 1–3), commonly known as Al(IV)-2, which usually resulted from synthesis and post-treatment processes [28,29,30,31]. Another signal at ∼0 ppm was ascribed to octahedrally coordinated Al (denoted as Al(VI)), namely extra-framework Al (EFAl) [32]. Acid etching dramatically changed the Al coordination environment of H-SSZ-39(OA), with Al(IV)-1 becoming the dominant component (∼48.7%) and the proportions of Al(IV)-2 and Al(VI) atoms decreased (Figure S14a). This provided direct evidence for the preferential interaction of OA with Al−OH groups and EFAl. For In/H-SSZ-39(OA), Al(IV)-2 again became the dominant component (∼36.3%), with a concomitant increase in Al(VI) and a decrease in Al(IV)-1. The migration of In species likely affected the zeolite pore structure that was accompanied by FAl-to-EFAl conversion [25]. In the ²⁹Si MAS SSNMR (Figure 4b), the signals at $-111$, $-105$, and $-99$ ppm could be assigned to Si(0Al), Si(1Al), and Si(2Al), respectively. In addition, a signal positioned between Si(1Al) and Si(2Al) was identified as Si−OH groups [31]. The ²⁹Si MAS SSNMR spectra were used to estimate the Si/Al ratio in the zeolite framework (abbreviated as Si/Al_f) [33] according to Equation (1). The calculated Si/Al_f followed a sequence of 13.8 for H-SSZ-39(OA) > 9.46 for In/H-SSZ-39(OA) > 8.5 for Pristine H-SSZ-39 (Figure 4b and S14b). The framework Si/Al_f ratios were higher compared to their bulk Si/Al ratios, also signifying the presence of EFAl, consistent with the ²⁷Al MAS NMR measurements. The increased Si/Al_f ratio after acid etching indicated the selective extraction of FAl by OA; whereas the decreased Si/Al_f ratio after In exchange could be related to the migration of In species.

2.2.5. Surface Chemical State

The chemical states of the catalyst surface components were examined using XPS. The In content on the In/H-SSZ-39(OA) surface measured by XPS (∼5.98 wt%, Figure S15a) was higher than that in the bulk phase determined by ICP (∼4.40 wt%, Table 1). For In-Co₃O₄/H-SSZ-39(OA), the surface enrichment of In and Co was also observed (Figure S15b and Table 1), consistent with the formation of In/Co oxides on the surface as observed by HRTEM. As shown in Figure 5a, In 3d spectral peaks of both In/H-SSZ-39(OA) and In-Co₃O₄/H-SSZ-39(OA) broadened with respect to those of reference In₂O₃, indicating the presence of a second In species that was commonly recognized as InO⁺ [25,34,35]. It was reported that InO⁺ species were the principal active centers in In-exchanged zeolitic CH₄-SCR catalysts responsible for CH₄ activation and active N_xO_y formation [15,25,36]. The percentage of InO⁺ species (InO⁺/In$\mathrm{all}$) in the In/H-SSZ-39(OA) catalyst was 5.7%. With Co₃O₄ modification, the percentage of InO⁺ species in In-Co₃O₄/H-SSZ-39(OA) increased to 12.6%, which could contribute to the high CH₄-SCR activity of the In-Co₃O₄/H-SSZ-39(OA) catalyst. Figure 5b depicted the high-resolution Co 2p spectra of reference Co₃O₄ and In-Co₃O₄/H-SSZ-39(OA). After deconvolution, it could be concluded that Co(II) oxide and Co(III) oxide coexisted on the In-Co₃O₄/H-SSZ-39(OA) surface according to the distinguishable characteristic satellite peaks of Co²+ (BE ≈ 787.0 eV) and Co³+ (BE ≈ 790.9 eV). Additionally, no detectable cobalt oxides other than Co₃O₄ were observed in the XRD analysis (Figure S13a), further indicating the presence of Co₃O₄ on the In-Co₃O₄/H-SSZ-39(OA) surface.

2.2.6. Surface Acidity

NH₃-TPD measurements were performed to assess the quantity and strength of surface acid sites of samples. As depicted in Figure 6a, the NH₃-TPD profile of the Pristine H-SSZ-39 could be deconvoluted with the Gaussian algorithm to four distinct NH₃ desorption peaks within the temperature range of 50–700 °C. Peak I (∼107 °C) was assigned to surface hydroxyl groups, such as Si−OH and Al−OH [25]. Peak II (∼172 °C) and Peak III (∼390 °C) were associated with the desorption of NH₃ bound to weak and strong Lewis acid sites (LAS, such as EFAl), respectively. Peak IV (∼536 °C) corresponded to NH₃ desorption from strong Brønsted acid sites (BAS, Si−OH−Al) [9,33,37,38]. In/H-SSZ-39(OA) exhibited a similar NH₃-TPD profile, but the total acid quantity (1.276 mmol g⁻¹) was reduced compared to H-SSZ-39 (1.366 mmol g⁻¹), as presented in

Table 2. The strength and quantity of surface acid sites of samples based on NH₃-TPD measurements.

Sample	Peak I		Peak II		Peak III		Peak IV
	T (°C)	Q (mmol g−1)	T (°C)	Q (mmol g−1)	T (°C)	Q (mmol g−1)	T (°C)	Q (mmol g−1)
Pristine H-SSZ-39	107.4	0.160	171.8	0.737	390.4	0.199	535.7	0.269
In/H-SSZ-39(OA)	107.2	0.141	168.1	0.615	370.3	0.301	530.1	0.219
In-Co3O4/H-SSZ-39(OA)	107.6	0.127	165.9	0.541	343.0	0.340	475.1 + 555.6	0.190 + 0.053

. Specifically, the quantity of weak LAS and strong BAS decreased, consistent with the expected results of acid etching [39]. The acid-etching dealumination might preferentially take place from EFAl compared to FAl, considering the presence of FAl-to-EFAl conversion during dealumination as well as the consumption of BAS during In exchange, which was consistent with the ²⁷Al MAS SSNMR results. On the other hand, the introduced In species apparently contributed to the increased strong LAS density. For In-Co₃O₄/H-SSZ-39(OA), the similar NH₃ desorption peaks could be observed at low temperatures (∼108 °C and ∼166 °C) with further decreased intensity, which might be associated with the partial coverage of the zeolite surface by In₂O₃ and Co₃O₄ nanoparticles. The peak assigned to NH₃ desorption from BAS, however, was split into two peaks (denoted as Peak IV and Peak IV’) in In-Co₃O₄/H-SSZ-39(OA); one was at ∼475 °C and the other one was at ∼556 °C. It was reported that the NH₃ desorption peak for pure Co₃O₄ was lower than 220 °C with a much smaller NH₃ desorption amount than those from parent zeolites or Co/zeolites [40]. The strong NH₃ desorption at ∼556 °C over In-Co₃O₄/H-SSZ-39(OA) suggested a probable synergistic interaction between Co species and the support acid sites. Similar phenomena had been observed in Co/Beta and Co/ZSM-5 zeolite [41].

2.2.7. N_xO_y Intermediates during Reaction

NO could be adsorbed on H-SSZ-39 in the form of N_xO_y species via the interaction with acidic hydroxyl groups in the zeolite [6]. Thus, NO-TPD was an effective tool to provide insights into possible surface N_xO_y species and their stability during SCR reactions over catalysts. As shown in Figure 6b, the vast majority of N_xO_y species escape from In/H-SSZ-39(OA) below 350 °C, with NO and N₂O detected as major desorption and/or decomposition products; whereas at higher temperatures, only a weak NO desorption peak centered at ∼580 °C was observed. Obviously, these weakly bound N_xO_y species with low decomposition temperatures could not persist at the active temperature (400–650 °C) for the CH₄-SCR reaction. Therefore, lacking available intermediate species, the CH₄-SCR activity of In/H-SSZ-39(OA) was relatively low. On the contrary, for In-Co₃O₄/H-SSZ-39(OA), both a weak NO desorption peak at ∼104 °C and a strong NO desorption peak at ∼525 °C were observed. In addition, N₂O was detected as a desorption and/or decomposition product at high temperatures, indicating that it might also be an available active N_xO_y species to serve as intermediates for CH₄-SCR. It was tentatively concluded that a small amount of Co₃O₄ rendered In-Co₃O₄/H-SSZ-39(OA) with more strong adsorption sites to enrich stable N_xO_y species reserves at high temperatures, thus boosting the CH₄-SCR activity of the catalyst.

3. Materials and Methods

3.1. Acid Etching

One gram of commercial H-SSZ-39 zeolite (Si/Al = 4.72, Dalian Huayizhongxin New Material Co., Ltd., Dalian, China) was ultrasonically dispersed in 25 mL of 1 M oxalic acid (oxalic acid dihydrate, AR, Damao Chemical Reagent Factory, Tianjin, China) aqueous solution. After stirring at 80 °C for 4 h, the etched H-SSZ-39 was recovered by suction filtration and washed with ultrapure water several times until the pH of the filtrate was approximately neutral. The filter cake was dried at 80 °C overnight followed by grinding and calcination in a muffle furnace at 500 °C for 3 h to obtain OA-etched H-SSZ-39, denoted as H-SSZ-39(OA).

3.2. Indium Exchange

A total of 0.3 g of H-SSZ-39(OA) zeolites were dispersed in 10 mL of 0.066 M indium nitrate (indium nitrate hydrate, 99.99% metals basis, Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China) aqueous solution and the suspension was stirred at 85 °C for 8 h. In-containing H-SSZ-39(OA) was recovered by centrifugation and washed twice with ultrapure water. After drying at 80 °C overnight, the sample was calcined in a muffle furnace at 500 °C for 3 h to obtain In/H-SSZ-39(OA). In-Co₃O₄/H-SSZ-39(OA) was prepared by the same process as that for In/H-SSZ-39(OA), except that a certain amount of Co₃O₄ fine powder (cobalt oxide, 99.9% metals basis, Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China) was suspended in the indium nitrate aqueous solution with the Co₃O₄: H-SSZ-39(OA) mass ratios of 1:5, 1:10, 1:20, 1: 30, 1:40, and 1:50.

3.3. Catalyst Characterizations

The microstructures of samples were observed with a JEOL JSM-7800F (Tokyo, Japan) field emission scanning electron microscope (FESEM) equipped with X-MaxN Falcon (Oxford Instruments, Abingdon, UK) energy dispersive X-ray spectroscopy (EDS) at 5 kV (15 kV for EDS mapping measurement) and a Talos F200X G2 (Thermo Fisher Scientific, Norristown, PA, USA) scanning transmission electron microscope ((S)TEM) equipped with Super-X G2 (Thermo Fisher Scientific) EDS at 200 kV. In, Co contents, and bulk Si/Al ratios of samples were measured using inductively coupled plasma optical emission spectroscopy (ICP-OES) on an Agilent 720ES (Santa Clara, CA, USA) instrument. N₂ adsorption–desorption isotherms measured at 77 K (Micromeritics ASAP2460, Norcross, GA, USA) were used to analyze specific surface areas and pore size distributions of the catalysts. The samples were degassed at 300 °C for 8 h before measurement. Powder X-ray diffraction (PXRD) patterns of samples were recorded by a Rigaku D/Max-2200 PC diffractometer (Tokyo, Japan) in the diffraction angle range of 2$\theta$ = 5–60° with Cu K$\alpha$ radiation ($\lambda$ = 1.5418 Å) at 40 kV and 50 mA. X-ray photoelectron spectra (XPS) were recorded by a Thermo Scientific K-Alpha instrument (America) with a monochromatic Al K$\mathsf{\alpha }$ (1486.6 eV) as an X-ray source. The binding energy values were calibrated using the C 1s peak at 284.8 eV for adventitious carbon. Solid-state nuclear magnetic resonance (SSNMR) experiments were performed on a Bruker 400M spectrometer (Bremen, Germany). The deconvolutions of spectra were performed with the ssNake v1.4 software [42]. The single-pulse ²⁹Si magic angle spinning (MAS) SSNMR spectra were acquired on a 7 mm probe with a spinning rate of 5 kHz, a pulse width of 5.6 $\mathsf{\mu }$s, a relaxation delay of 5 s, and 256 scans. The ²⁹Si chemical shifts were referenced to tetramethylsilane (TMS) at 0 ppm, and the framework Si/Al_f ratios were estimated using the following Equation (1).

$$\mathrm{Si}/{\mathrm{Al}}_{\mathrm{f}}={\displaystyle \frac{{\sum }_{\mathit{n}=0}^4{\mathit{I}}_{\mathrm{Si}\left(\mathit{n}\mathrm{Al}\right)}}{{\sum }_{\mathit{n}=0}^{4}0.25\mathit{n}\left[{\mathit{I}}_{\mathrm{Si}\left(\mathit{n}\mathrm{Al}\right)}\right]}}$$

(1)

where $I_{\mathrm{Si}(\mathit{n}\mathrm{Al})}$ was the signal intensity of Si with different numbers of incorporated Al atoms (n = 0–4) in its first Si(OT)₄ (T representing framework Si and Al atoms) coordination shell. The single-pulse ²⁷Al MAS SSNMR were acquired on a 4 mm probe with a spinning rate of 10 kHz, a pulse width of 1.48 $\mathsf{\mu }$s, a relaxation delay of 0.1 s, and 2048 scans. The ²⁷Al chemical shifts were referenced to NaAlO₂ at 65 ppm. Ammonia temperature-programmed desorption (NH₃-TPD) was carried out on a Micromeritics AutoChem II 2920 chemisorption analyzer (America). A total of 100 mg of sample was loaded into a quartz reactor and pretreated in air at 500 °C for 60 min. After cooling to 50 °C, 10% NH₃/He was introduced at a flow rate of 50 mL min⁻¹ for 60 min to saturate the sample with NH₃, followed by introducing He flow as purge gas. The NH₃-TPD profile was then recorded by a thermal conductivity detector (TCD) across the temperature range of 50 °C to 700 °C at a ramping rate of 10 °C min⁻¹. Nitric oxide temperature-programmed desorption (NO-TPD) analysis was performed on a TP-5080-B instrument (Tianjin, China) equipped with a Hiden DECRA mass spectrometer (Warrington, UK). A total of 100 mg of sample was loaded into a quartz reactor and pretreated in a He stream (30 mL min⁻¹) at 500 °C for 60 min. After cooling to 50 °C, the He flow was switched to a 10% NO/He gas mixture at a flow rate of 30 mL min⁻¹. The weakly physically adsorbed NO was removed by purging with He gas flow (30 mL min⁻¹) for 60 min before programmed warming at 10 °C min⁻¹. The following mass fragments sensible to the system perturbation were monitored on-line in the temperature range of 50–660 °C: NO (m/z = 30), O₂ (m/z = 32), N₂O (m/z = 44), and NO₂ (m/z = 46).

3.4. Catalytic Activity Test

CH₄-SCR activity was tested at atmospheric pressure using a certain mass of 40–60 mesh catalyst loaded in a fixed-bed quartz reactor [21]. A gas mixture composed of 400 ppm CH₄, 600 ppm NO, 10 vol% O₂, and 5 vol% H₂O (optional) with an Ar balance at a flow rate of 100 mL min⁻¹ was introduced into the reactor, corresponding to a gas hourly space velocity (GHSV) of ∼24,000 h⁻¹ for 0.1 g of catalyst. The concentrations of NO_x were monitored by a nitrogen oxide analyzer (Teledyne Model T200H, Thousand Oaks, CA, USA), while CH₄, CO, and CO₂ concentrations were analyzed by a gas chromatograph (Fuli GC9790II, Taizhou, China) equipped with a Porapak-Q column (Agilent, America) and a flame ionization detector (FID). The NO_x and CH₄ conversions as well as CH₄ selectivity were calculated using Equations (2)–(4), respectively.

$${\mathrm{NO}}_{\mathit{x}}\mathrm{conversion}(\%)={\displaystyle \frac{{\left[{\mathrm{NO}}_{\mathit{x}}\right]}_{\mathrm{in}}-{\left[{\mathrm{NO}}_{\mathit{x}}\right]}_{\mathrm{out}}}{{\left[{\mathrm{NO}}_{\mathit{x}}\right]}_{\mathrm{in}}}}\times 100\%$$

(2)

$${\mathrm{CH}}_4\mathrm{conversion}(\%)={\displaystyle \frac{{\left[{\mathrm{CH}}_4\right]}_{\mathrm{in}}-{\left[{\mathrm{CH}}_4\right]}_{\mathrm{out}}}{{\left[{\mathrm{CH}}_4\right]}_{\mathrm{in}}}}\times 100\%$$

(3)

$${\mathrm{CH}}_4\mathrm{selectivity}(\%)=0.5\times {\displaystyle \frac{{\left[{\mathrm{NO}}_{\mathit{x}}\right]}_{\mathrm{in}}-{\left[{\mathrm{NO}}_{\mathit{x}}\right]}_{\mathrm{out}}}{{\left[{\mathrm{CH}}_4\right]}_{\mathrm{in}}-{\left[{\mathrm{CH}}_4\right]}_{\mathrm{out}}}}\times 100\%$$

(4)

where NO_x represents NO and NO₂; the subscripts “in” and “out” represent inlet and outlet, respectively.

4. Conclusions

In summary, In-Co₃O₄/H-SSZ-39(OA) has been successfully constructed and applied as a robust catalyst in CH₄-SCR reaction under harsh conditions. Specifically, the NO conversion of ∼80% could be achieved at ∼600 °C under a GHSV of 24,000 h⁻¹ and in the presence of 5 vol% H₂O, significantly outperforming the In/H-SSZ-39 without acid etching pretreatment and the In/H-SSZ-39(OA) without Co₃O₄ modification. Moreover, good stability was achieved on In-Co₃O₄/H-SSZ-39(OA) and <15% activity loss could be observed within 7 h at 600 °C. The mild acid leaching with OA delicately tuned the Si/Al ratio of SSZ-39 zeolite. During this process, OA was found to preferentially interact with Al−OH and EFAl, with some mesopores introduced while maintaining relative high crystallinity. A small amount of Co₃O₄ greatly improved the catalytic activity of the catalyst despite causing the severely decreased the crystallinity of SSZ-39 zeolite. Co₃O₄ could promote CH₄ conversion and render a much higher storage of stable N_xO_y species available at high temperatures.