Nitrogen oxides derived from boilers, engines, and power plants contribute to the formation of photochemical smog and acid rain and are harmful to human health [1]. The selective catalytic reduction of nitrogen oxides (NO and NO₂) by hydrocarbons (HC-SCR) is a promising strategy for the post-treatment of nitrogen oxides in excess oxygen, and CH₄ is an attractive reductant because of its low cost and easy availability in natural gas power plants. In China, CH₄-SCR is now attracting particular interest in the move to replace coal with natural gas as a raw material for urban power plants. However, because of its chemical inertness, methane is difficult to activate, which remains a key problem in the CH₄-SCR [2-5]. However, NO oxidation, the role of which is still unclear and seems dependent on the catalysts employed [6], is an important step in the CH₄-SCR.

Zeolites, known as microporous crystalline aluminosilicates, can be directly used as catalysts [7, 8] and are more frequently used as catalyst supports [9] in SCR. Transition-metal-modified zeolites have been extensively investigated for CH₄-SCR in past decades [10-33]. Among them, In-modified zeolites appear to be most active monometallic catalysts for CH₄-SCR because of the effective activation of CH₄ by In species [22-29, 32]. It was claimed that methane can be activated on In₂O₃ sites to generate oxygenates, which then react with nitrate to generate N₂ [22]. Alternatively, the H₂NCO intermediate formed by the reaction between NO₂ and partially oxidized methane on intrazeolite InO⁺ sites was proposed to be the actual NO reductant in the CH₄-SCR [23]. To improve the CH₄-SCR activity, second transition metals, e.g., Pd [13, 14], Co [3, 19], and Ce [20], were introduced to In-zeolites, and the major role of the second transition metals was proposed to promote the oxidative activation of NO [3, 14, 16, 20]. For example, the cobalt oxide clusters in Co-In/HZSM-5 showed a positive effect on the oxidation of NO to NO₂, which promoted the CH₄-SCR [3]. Similarly, palladium in Pd-In/H-ZSM-5 promoted the oxidation of NO and increased the formation of the activated nitrate species, while In⁺/InO⁺ sites suppressed the formation of less reactive isocyanate and nitrile species [14].

In this work, we aimed to develop an efficient catalyst for the CH₄-SCR, i.e., achieving good activity and stability in the presence of excess H₂O and under high space velocity conditions. H-SSZ-13, a high-silica zeolite with a CHA topology, was first used as a catalytic support for CH₄-SCR because of its high stability against framework dealumination. Bimetallic Cr-In/H-SSZ-13 was optimized and its structure-activity relationship in the CH₄-SCR was analyzed.

All the chemical reagents employed in this study were of analytical grade from Alfa Aesar and used as received without further purification. Commercial zeolites in their H form with similar Si/Al ratios of 24, i.e., H-SSZ-13, H-ZSM-5, and H-beta, as well as amorphous SiO₂ (surface area of 210 m²/g) were used as supports, and metal modifiers were introduced via wet impregnation. In a typical process, the zeolite support was immersed in a solution containing the desired amount of indium nitrate and chromium nitrate and stirred at room temperature for 24 h. Subsequently, the solvent of the slurry was removed in a rotary evaporator at 80 ℃, and the residue was dried in an oven at 80 ℃ for 12 h. The obtained solid sample was calcined in Ar at 550 ℃ for 2 h, reduced in 10% H₂/Ar at 450 ℃ for 1 h, and oxidized in 10% O₂/Ar at 450 ℃ for 1 h. The final product was denoted as x%Cr-y%In/Z, where x% and y% indicate the weight loadings of Cr and In, respectively, and Z indicated the type of zeolite support. Bimetallic In-containing samples, i.e., Me-In/H-SSZ-13 (Me = Ti, V, Mn, Fe, Co, Ce, Zr, and Mo), were prepared via similar procedures.

The chemical compositions of samples were analyzed on an IRIS Advantage inductively coupled plasma atomic emission spectrometer.

Transmission electron microscopy (TEM) images of selected samples were acquired on an FEI Tecnai G² F20 electron microscope. High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images were acquired on an FEI Talos electron microscope. Element mapping analysis was conducted under HAADF-STEM mode using an FEI built-in energy dispersive spectrum.

X-ray photoelectron spectra (XPS) of samples were acquired on a Thermo Scientific ESCALAB 250Xi spectrometer with a monochromatic Al K_α X-ray source (hν = 1486.6 eV). Accurate binding energies (±0.1 eV) were determined with reference to the C 1s line of adventitious carbon at 284.8 eV.

The temperature-programmed desorption of ammonia (NH₃-TPD) was performed on a Quantachrome ChemBET 3000 chemisorption analyzer. In a typical experiment, the sample was saturated with 5% NH₃/He at 50 ℃ and then purged with He at the same temperature for 1 h to eliminate the physical absorbed ammonia. The NH₃-TPD profile was recorded in flowing He at a heating rate of 10 ℃/min from 50 to 600 ℃.

The experiments of temperature-programmed reduction by hydrogen (H₂-TPR) were also performed on the Quantachrome ChemBET 3000 chemisorption analyzer. In a typical experiment, a sample of 0.1 g was pretreated in 10% O₂/He at 450 ℃ for 1 h, cooled to 50 ℃, and purged for 1 h in flowing He. The H₂-TPR profile was recorded in 5% H₂/Ar (30 mL/min) at a heating rate of 10 ℃/min. The outlet gas was passed through a dry-ice trap, and the hydrogen consumption was calculated using CuO as a reference.

The CH₄-SCR reaction was performed in a fixed-bed micro-reactor at atmospheric pressure. Typically, a 0.12 mL catalyst sample (20–40 mesh) was placed in a quartz reactor and pretreated in 10% O₂/Ar at 450 ℃ for 1 h. After cooling to the designated temperature in He, the reactant gas mixture (NO = 2500 ppm; CH₄ = 4000 ppm; O₂ = 4%, H₂O = 6%, He balance) was fed to the catalyst to start the reaction. The total flow rate was set at 150 mL/min resulting in a gas hourly space velocity (GHSV) of 75 000 /h. The outlet gas (H₂O removed by cold trap) was analyzed on-line by a NOx analyzer (Ecotech EC9841) and a gas chromatograph (Techcomp GC7900, equipped with a Plot TDX-1 packed column and an FID detector for the analysis of CH₄ and CO_x, as well as Porapak Q packed column and TCD detector for the analysis of N₂O and N₂). During the reaction, the outlet gas stream was analyzed by a mass spectrometer (Pfeiffer Omnistar GSD 320), and the following mass fragments sensible to the system perturbation were monitored: CH₄ (m/e = 15), NO/NO₂ (m/e = 30), NO₂ (m/e = 46), O₂ (m/e = 32), N₂ (m/e = 28), CO₂ (m/e = 44), NH₃ (m/e = 17), H₂O (m/e = 18), and HCHO (m/e = 29). The NO and CH₄ conversions are defined as follows:

The temperature-programmed surface reaction (TPSR) of CH₄-SCR was performed on a quartz tube reactor, and the products were analyzed on-line by a Pfeiffer Omnistar GSD 320 mass spectrometer.

In the representative TEM images of 0.5%Cr/H-SSZ-13 (Fig. 1(a)) and 2%In/H-SSZ-13 (Fig. 1(b)), irregular polygon-like particles with sizes of a dozen nanometers corresponding to the Cr-and In-containing species on the surface of H-SSZ-13 zeolite support can be observed. In contrast, much smaller uniform nanoparticles were distributed on zeolite in the case of 0.5%Cr-2%In/H-SSZ-13 (Fig. 1(c)). From the HAADF-STEM element mapping images (Fig. 1(d) and (e)), the nanoscale aggregations of the Cr and In species can be clearly identified. Meanwhile, the Cr and In species seem to be located in the same regions (marked with white dotted circles), revealing the intimate contact between the Cr and In species. This is further confirmed by a line-scan analysis (red line 1 in Fig. 1(d)) where both the Cr and In species were detected in a single nanoparticle. Since the 0.5%Cr-2%In/H-SSZ-13 sample was submitted to reduction-oxidation treatments at elevated temperatures (see experiment section for details), the Cr and In species should exist in the thermodynamically stable states on H-SSZ-13 support, and the intimate contact between the Cr and In species should originate from their intrinsic properties.

The acidic properties of zeolite catalysts were characterized by means of NH₃-TPD, and the results are shown in Fig. S1. For H-SSZ-13 support, two distinct ammonia desorption peaks were observed corresponding to weak acid sites (peak centered at 200 ℃) and strong acid sites (peak centered at 450 ℃). The introduction of Cr and/or In to H-SSZ-13 resulted in noticeable decreases in the intensity of strong acid sites because of the interaction of Cr and/or In species with the Br nsted acid sites. Nevertheless, the strong acid sites in parent SSZ-13 zeolite were well preserved after Cr and/or In impregnation and can participate in both methane activation [34] and NO activation [9, 14] for CH₄-SCR.

The chemical states of Cr and In species on the zeolite support were investigated by means of XPS. For 0.5%Cr/H-SSZ-13, binding energy peaks at 586.9 (2p_1/2) and 577.4 (2p_3/2) eV were observed (Fig. 2(a)) corresponding to the mixture of Cr³⁺/Cr⁶⁺ [35, 36]. The deconvolution of these peaks was difficult because of the poor signals of Cr with low loading of 0.5%. For 0.5%Cr-2%In/H-SSZ-13, similar binding energy peaks of Cr³⁺/Cr⁶⁺ were observed. The difference in the chemical states of Cr in 0.5%Cr/H-SSZ-13 and 0.5%Cr-2%In/H-SSZ-13, even if existent, could not be distinguished through XPS analysis. In the In 3d XPS, peaks at 453.9 and 446.5 eV were observed for 2%In/H-SSZ-13 attributed to the 3d_3/2 and 3d_5/2 of (InO)⁺ and/or In₂O₃ species interacting with the zeolite support, respectively [18, 37]. For 0.5%Cr-2%In/H-SSZ-13, two additional binding energy value signals at 452.5 and 445.0 eV were clearly observed (Fig. 2(b)). Since the sample was finally oxidized in 10% O₂/Ar at 450 ℃, the presence of low-valence In species can be ruled out. In this context, these binding energies should be related to the oxidized In species interacting with the higher-charge density of Cr³⁺/Cr⁶⁺. That is, the electron interaction between the Cr and In species can be revealed by XPS analysis.

The chemical states of the Cr and In species on the zeolite supports were further examined by TPR, and the H₂-TPR profiles are shown in Fig. 3. For 0.5%Cr/H-SSZ-13 (Fig. 3(b)), a series of reduction peaks in the temperature region 250–550 ℃ were observed corresponding to the reduction of Cr⁶⁺ to Cr³⁺ and/or Cr³⁺ to Cr²⁺ in different chemical environments [35, 38]. The calculated H/Cr ratio was 1.6, confirming the XPS results that a mixture of Cr³⁺/Cr⁶⁺ species existed in 0.5%Cr/H-SSZ-13 (Fig. 2). For 2%In/H-SSZ-13 (Fig. 3(a)), a single characteristic reduction peak at 290 ℃ with a H/In ratio of 1.9 attributed to the reduction of In³⁺ to In⁺ was observed [26]. In the case of bimetallic 0.5%Cr-2%In/H-SSZ-13 (Fig. 3(c)), the reduction of the Cr and In species was well defined: the reduction of the In species was not distinctly affected by the presence of Cr (H/In = 1.8, reduction peak at 290 ℃), and the reduction of the Cr species was affected by the presence of In (H/Cr = 1.6, two reduction peaks at 360 and 440 ℃). The similar H/In and H/Cr ratios observed for bimetallic 0.5%Cr-2%In/H-SSZ-13 reveal the similar overall average oxidation states of the Cr and In species on the zeolite support with reference to monometallic samples, while the noticeable changes in the reduction peaks corresponding to the Cr species confirms the electron interaction between Cr and In as also revealed by XPS (Fig. 2).

The catalytic performance of 0.5%Cr/H-SSZ-13, 2%In/H-SSZ-13 (optimized In loading employed), and 0.5%Cr-2%In/H-SSZ-13 in the CH₄-SCR reaction was investigated. As shown in Fig. 4, 0.5%Cr/H-SSZ-13 (Fig. 4(a)) exhibited very low activity in the CH₄-SCR reaction (NO conversion to N₂ < 20% at 550 ℃), and 2%In/H-SSZ-13 (Fig. 4(b)) exhibited considerable activity (NO conversion to N₂ > 40% at 550 ℃). The physical mixture of 0.5%Cr/H-SSZ-13 (0.12 mL) and 2%In/H-SSZ-13 (0.12 mL) (Fig. 4(c)) appeared to be more active, but it acted more like a simple accumulation of two individual components. In contrast, bimetallic 0.5%Cr-2%In/ H-SSZ-13 (Fig. 4(d)) was very active in the CH₄-SCR reaction. Typically, NO conversion increased from 10% to more than 90% with an increasing reaction temperature from 300 to 550 ℃, and N₂ selectivity greater than 99% was obtained at temperatures greater than 450 ℃. Obviously, a cooperative effect between the Cr and In species did exist and played a key role in the CH₄-SCR reaction catalyzed by bimetallic 0.5%Cr-2%In/ H-SSZ-13. In other words, the presence of Cr dramatically promoted the CH₄-SCR reaction over In/H-SSZ-13.

The catalytic performance of bimetallic Cr-In on different supports was investigated. As shown in Fig. 5, the support materials played a decisive role in the CH₄-SCR activity of the Cr-In catalysts. H-SSZ-13 appears to be the best support for the Cr-In active species, followed by H-ZSM-5. H-beta and SiO₂ are not suitable support materials, and very low CH₄-SCR activity was observed for Cr-In/H-beta and Cr-In/SiO₂. H-SSZ-13 can be clearly optimized from all the support materials investigated. An extra advantage of using H-SSZ-13 as support is its remarkable thermal and hydrothermal stability, which should be good for the catalytic stability of 0.5%Cr-2%In/H-SSZ-13 at elevated temperatures and in the presence of excess H₂O (vide infra).

A series of bimetallic In-containing zeolite catalysts were further applied in the CH₄-SCR reaction to elaborate the unique promotion effect of the Cr species. As shown in Fig. 6, some elements, i.e., Fe, V, Mo, Cr, Ce, and Co show promotion effects on In/H-SSZ-13 for the CH₄-SCR reaction, while others (Ti, Mn, and Zr) do not. It has been reported that Ce [20] and Co [3, 19] can significantly promote the CH₄-SCR reaction over In/zeolites, and similar promotion effects were observed here. Fortunately, the best promotion effect of Cr on In/H-SSZ-13 for CH₄-SCR was disclosed for the first time in this study. To our knowledge, bimetallic 0.5%Cr-2%In/H-SSZ-13 appears to be the most active CH₄-SCR catalysts under comparable reaction conditions, i.e., high NO inlet concentration, high GHSV, and excessive H₂O [3, 10, 14, 30, 39]. For bimetallic x%Cr-2%In/ H-SSZ-13 catalysts, increasing the Cr loading from 0.2% to 0.5% showed noticeable positive impacts on the CH₄-SCR activity, while further increases of Cr loading to 1% had no significant effect (Fig. S2).

Some reaction parameters, e.g., space velocity (Fig. S3) and O₂ and CH₄ concentration in feeding gas (Fig. S4), were also investigated. In the absence of O₂, relatively low NO conversion to N₂ was observed above 300 ℃, which should come from the direct decomposition of NO (Eq. (4)). The introduction of O₂ to the reaction system can help to activate NO and/or CH₄, and, therefore, trigger the CH₄-SCR reaction [39, 40]. More O₂ can slightly promote NO oxidation below 300 ℃ and CH₄ oxidation at temperatures greater than 400 ℃. In this study, excess CH₄ was employed for NO (Eq. (1)) since CH₄ is cheap and easily available in natural gas power plants and the unreacted CH₄ can be removed by catalytic combustion (Eqs. (2) and (3)). As expected, changing the CH₄ concentration within the range 3000–5000 ppm did not significantly influence the CH₄-SCR activity, while the efficiency of the CH₄ reducing agent differed slightly (Fig. S4).

According to our experimental observations and literature reports [4, 23], the possible reactions during CH₄-SCR are identified below:

(1)

(2)

(3)

(4)

(5)

(6)

Among these reactions, CH₄ oxidation to CO₂ or CO by O₂ (Eqs. (2) and (3)) and NO oxidation to NO₂ by O₂ (Eq. (5)) are known as major side reactions. However, they are related to the activation of CH₄ and NO, respectively, and show significant impacts on CH₄-SCR. Therefore, the reactions catalyzed by 0.5%Cr/H-SSZ-13, 2%In/H-SSZ-13, and 0.5%Cr-2%In/ H-SSZ-13 were investigated. As shown in Fig. S5, noticeable NO oxidation to NO₂ with maximal NO conversion of 25% was achieved in the temperature range 200–450 ℃ over all three catalysts. Obviously, Cr does not significantly promote the oxidation of NO to gaseous NO₂. Moreover, the gas-phase NO₂ formation is not the rate-determining step in CH₄-SCR since the CH₄-SCR rate is apparently higher than that of NO oxidation (Figs. 4 and S5).

For CH₄ oxidation, 0.5%Cr/H-SSZ-13 exhibited relatively low activity and a considerable amount of CO was produced from incomplete oxidation (Fig. S6). 2%In/H-SSZ-13 exhibited moderate activity in CH₄ oxidation, with CO₂ as the dominating product. 0.5%Cr-2%In/H-SSZ-13 appeared to be more active than 0.5%Cr/H-SSZ-13 and 2%In/H-SSZ-13, and the formation of CO as observed in the case of 0.5%Cr/H-SSZ-13 was significantly suppressed. These observations indicate the cooperation between Cr and In during CH₄ oxidation. It has been claimed that CH₄ oxidation competes with CH₄-SCR, especially at high temperatures, and higher CH₄ oxidation activity may lead to lower CH₄-SCR activity [41]. However, CH₄ oxidation activity may indicate the capability of CH₄ activation, which is also very important for CH₄-SCR. Since excess CH₄ was employed in the CH₄-SCR in this study, higher CH₄ oxidation activity correlates with higher CH₄-SCR activity (Figs. 4 and S6).

The stability of 0.5%Cr-2%In/H-SSZ-13 during CH₄-SCR at 500 ℃ was investigated. As shown in Fig. 7, NO conversion to N₂ over 0.5%Cr-2%In/H-SSZ-13 was very stable for the first 30 h of the CH₄-SCR reaction in the absence of H₂O (NO conversion of 84%). The introduction of 6% H₂O to the reaction system caused a slight decrease in NO conversion to 82%, which was stable for the next 90 h. After the feeding of H₂O was stopped, the NO conversion gradually recovered to the initial value and remained at this level for a further 30 h. These observations indicate that the presence of excess H₂O shows reversible negative impacts on CH₄-SCR, probably by competing adsorption on the active sites. The remarkable catalytic stability of 0.5%Cr-2%In/H-SSZ-13 can be explained from the stability of both the bimetallic active centers and the zeolite support. Through reduction-oxidation treatment, bimetallic Cr-In species exist in the thermodynamically stable states and their structures during reaction will not change. Meanwhile, H-SSZ-13 zeolite was very stable upon thermal and hydrothermal treatments, and the structure destruction because of the framework dealumination was completely suppressed. On the whole, 0.5%Cr-2%In/H-SSZ-13 appears to be an eligible CH₄-SCR catalyst for potential applications.

The reaction intermediates in CH₄-SCR were investigated by means of TPSR, and the results are shown in Fig. 8. The activity difference in CH₄-SCR was well identified and in good agreement with the steady-state reaction results (Fig. 4). Formaldehyde (HCHO) from the incomplete oxidation of methane was detected as the exclusive carbon-containing intermediate gaseous product during CH₄-SCR over 2%In/H-SSZ-13, indicating the oxidative activation of CH₄ to HCHO over the In species. The further oxidation of HCHO (by O₂, NO, NO₂, or other oxidants) to H₂O and CO₂ was apparently a slow step over 2%In/H-SSZ-13. While in the case of 0.5%Cr-2%In/H-SSZ-13, the formation of HCHO was greatly suppressed and a rational explanation is that the further oxidation of HCHO was promoted by the presence of the Cr species. Another interesting phenomenon is that the conversion of NO/NO₂ to N₂ was faster than the CH₄ conversion to CO₂ over 2%In/H-SSZ-13 (Fig. 8(b)), hinting to a mechanism of NO decomposition followed by oxygen removal. In contrast, a synchronous conversion of NO/NO₂ and CH₄ was observed over bimetallic 0.5%Cr-2%In/H-SSZ-13 (Fig. 8(c)). According to the TPSR results, we propose that the activation of NO occurs on Cr species, while the activation of CH₄ occurs on In species. The cooperation between Cr and In species is responsible for the remarkable CH₄-SCR activity of 0.5%Cr-2%In/H-SSZ-13 in accordance with the steady-state reaction results.

Bimetallic 0.5%Cr-2%In/H-SSZ-13 was successfully developed as a robust catalyst for CH₄-SCR, outperforming monometallic Cr/H-SSZ-13, In/H-SSZ-13, and their mixture. Typically, NO conversion greater than 90% with N₂ selectivity greater than 99% was achieved at 550 ℃ and high GHSV of 75 000 /h in the presence of 6% H₂O. Through reduction-oxidation treatments, Cr and In species existed in the thermodynamically stable states on H-SSZ-13. TEM images and XPS results revealed a close contact between Cr and In species as well as their interaction, which is helpful for constructing an efficient cooperative catalysis system. According to the catalytic data, the activation of CH₄ and NO occurred on the In and Cr sites of bimetallic Cr-In/H-SSZ-13, respectively, both with the participation of O₂.

More characterization results and catalytic results in CH₄-SCR.