Ag-Cu Synergism-Driven Oxygen Structure Modulation Promotes Low-Temperature NO_x and CO Abatement

Abstract

The efficient simultaneous removal of NO_x and CO from sintering flue gas under low-temperature conditions (110–180 °C) in iron and steel enterprises remains a significant challenge in the field of environmental catalysis. In this study, we present an innovative strategy to enhance the performance of CuSmTi catalysts through silver modification, yielding a bifunctional system capable of oxygen structure regulation and demonstrating superior activity for the combined NH₃-SCR and CO oxidation reactions under low-temperature, oxygen-rich conditions. The modified AgCuSmTi catalyst achieves complete NO conversion at 150 °C, representing a 50 °C reduction compared to the unmodified CuSmTi catalyst (T_100% = 200 °C). Moreover, the catalyst exhibits over 90% N₂ selectivity across a broad temperature range of 150–300 °C, while achieving full CO oxidation at 175 °C. A series of characterization techniques, including XRD, Raman spectroscopy, N₂ adsorption, XPS, and O₂-TPD, were employed to elucidate the Ag-Cu interaction. These modifications effectively optimize the surface physical structure, modulate the distribution of acid sites, increase the proportion of Lewis acid sites, and enhance the activity of lattice oxygen species. As a result, they effectively promote the adsorption and activation of reactants, as well as electron transfer between active species, thereby significantly enhancing the low-temperature performance of the catalyst. Furthermore, in situ DRIFTS investigations reveal the reaction mechanisms involved in NH₃-SCR and CO oxidation over the Ag-modified CuSmTi catalyst. The NH₃-SCR process predominantly follows the L-H mechanism, with partial contribution from the E-R mechanism, whereas CO oxidation proceeds via the MvK mechanism. This work demonstrates that Ag modification is an effective approach for enhancing the low-temperature performance of CuSmTi-based catalysts, offering a promising technical solution for the simultaneous control of NO_x and CO emissions in industrial flue gases.

1. Introduction

Flue gas generated during the fossil fuel sintering process in iron and steel enterprises contains harmful atmospheric pollutants, such as nitrogen oxides (NO_x) and carbon monoxide (CO) [1], which pose serious threats to both environmental quality and human health [2,3]. Among various NO_x removal technologies, ammonia-based selective catalytic reduction (NH₃-SCR) is widely recognized as the most effective denitrification method [4]. The V₂O₅-WO₃/TiO₂ catalyst, known for its excellent catalytic performance, has been extensively applied in industrial NH₃-SCR processes [5]. However, this catalyst operates optimally within a temperature range of 320–450 °C and involves the use of V₂O₅, which becomes highly toxic at elevated temperatures [6]. As a result, its application faces significant challenges when treating sintering flue gas at much lower temperatures (110–180 °C) [7]. In light of increasingly stringent environmental regulations, the development of high-performance denitrification catalysts suitable for low-temperature conditions has become an urgent priority. Meanwhile, catalytic CO oxidation, as a well-established technology for CO control, has been widely studied and implemented [8]. Nevertheless, the design and development of a bifunctional catalyst capable of simultaneously removing both NO_x and CO remains a key research focus and challenge in the field of environmental catalysis.

In theory, the selective catalytic reduction of NO_x using CO as a reductant (CO-SCR) offers a promising strategy for addressing the simultaneous removal of NO_x and CO [9]. However, in practical applications, the presence of a large excess of oxygen in flue gas causes CO to preferentially react with O₂ rather than NO_x [10], significantly limiting the effectiveness and applicability of the CO-SCR process. A promising alternative is the development of bifunctional catalysts capable of simultaneously promoting both CO oxidation and NH₃-SCR reactions under oxygen-rich conditions. These catalysts can utilize the heat generated during CO oxidation to enhance the performance of the NH₃-SCR reaction, thereby improving overall energy efficiency [11]. This dual-function strategy not only reduces energy consumption and operational costs but also provides a feasible technical pathway for the synergistic control of NO_x and CO emissions in low-temperature flue gas environments.

In recent years, transition metal catalysts such as Cu, Mn, and Fe have demonstrated significant potential in NH₃-SCR coupled with CO oxidation due to their excellent catalytic activity, superior thermal stability, and relatively low cost [12]. Among these, copper-based catalysts exhibit particular potential, owing to their multiple redox states and strong redox capabilities, which confer significant advantages in both NH₃-SCR [13] and CO oxidation [14] reactions. To further enhance the catalytic performance of copper-based catalysts, extensive modification and optimization efforts have been undertaken by researchers. Liu et al. [15] introduced phosphorus into a CuCeZr (CCZ) catalyst to achieve a balanced combination of acidity and redox properties, significantly enhancing the catalytic activity of the CCZ-P1 catalyst within the temperature range of 250–350 °C. Shen et al. [16] synthesized a molybdenum-doped CuCeO_x bifunctional catalyst and found that Mo incorporation enhanced the catalyst’s reducibility, promoted the adsorption and activation of reactant gases, and enabled the CCM3 catalyst to achieve a NO_x conversion rate of 91.2% at 225 °C, along with complete CO oxidation over a broad temperature window of 150–300 °C. Lv et al. [17] modified a Cu/TiO₂ catalyst with Sm, constructing asymmetric active sites of Cu⁺-Sm³⁺-O_v-Ti⁴⁺, which effectively facilitated full NO conversion at 200 °C and efficient CO oxidation in the temperature range of 175–300 °C.

Although modified copper-based catalysts such as CuSmTi have exhibited relatively superior low-temperature activity, their performance still does not fully meet the practical requirements for the treatment of sintering flue gas at temperatures below 180 °C. Therefore, further modification strategies are required to enhance their catalytic performance at such conditions. Liu et al. [18] prepared Ag-modified α-MnO₂ catalysts for efficient benzene oxidation. The Ag modification enriched the variety of oxygen species and increased the quantity and enhanced the reactivity of lattice oxygen, thereby promoting the reaction efficiency. Similarly, Rao et al. [19] reported that the incorporation of Ag could activate a substantial amount of lattice oxygen in CeO₂ nanoparticles. The lattice oxygen activated at the Ag-CeO₂ interface significantly improved the catalytic performance. These studies demonstrate that Ag modification plays a crucial role in regulating the structure and distribution of oxygen species, thereby effectively enhancing the overall catalytic performance of the material.

In this study, an Ag-modified AgCuSmTi bifunctional catalyst was successfully synthesized using the sol–gel method, which facilitated effective oxygen structure regulation. The catalyst exhibits excellent NO and CO conversion efficiencies at low temperatures while maintaining favorable N₂ selectivity. A suite of characterization techniques was employed to investigate the surface structure and internal chemical valence states, revealing pronounced Ag-Cu interactions within the AgCuSmTi catalyst. Further analysis indicated that these Ag-Cu interactions promote the optimization of oxygen structure and modulate the distribution of surface acid sites, thereby enhancing the low-temperature activity of the AgCuSmTi bifunctional catalyst. This study provides new insights into catalytic materials and a theoretical basis for the synergistic removal of multiple pollutants in sintering flue gas.

2. Results and Discussion

2.1. Evaluation of Catalytic Performance

To regulate the performance of the CuSmTi catalyst in the NH₃-SCR coupled with CO oxidation reaction, Ag species were introduced as dopants, and the influence of varying Ag doping levels on catalytic activity was systematically investigated. The catalytic performance of Ag_xCuSmTi catalysts with different Ag contents (x = 0.5, 0.8, 1.0, 1.2, 1.5) was evaluated to determine the optimal doping ratio, where x denotes the molar ratio of AgNO₃ to Cu(NO₃)₂·3H₂O. As shown in Figure S1a, the NO_x conversion at 150 °C gradually increases with increasing Ag content, reaching 100% for the Ag_1.2CuSmTi catalyst. However, further increasing the Ag doping ratio to x = 1.5 results in a slight decline in catalytic activity. Figure S1b presents the variation in N₂ selectivity with Ag loading. With increasing Ag content, the N₂ selectivity decreases gradually, and this trend becomes more pronounced at higher doping levels. Nevertheless, the Ag_1.2CuSmTi catalyst still maintains a N₂ selectivity exceeding 90%, demonstrating a favorable combination of high activity and selectivity. As depicted in Figure S1c, all catalysts achieve complete CO conversion at 175 °C. Based on the overall performance analysis, the Ag_1.2CuSmTi catalyst exhibits the best catalytic performance among the series. For simplicity and consistency in subsequent discussions, the Ag_1.2CuSmTi catalyst is hereafter referred to as AgCuSmTi.

After determining the optimal silver doping ratio, the effect of Ag modification on catalytic performance was further investigated by comparing the NH₃-SCR and CO oxidation activities of AgCuSmTi, CuSmTi, and AgSmTi catalysts. Figure 1a presents the NO_x conversion profiles of the three catalysts. The CuSmTi catalyst achieves complete NO_x conversion at 200 °C; however, its activity declines significantly at temperatures above 250 °C. Notably, the introduction of Ag markedly enhances low-temperature performance, enabling the AgCuSmTi catalyst to reach 100% NO_x conversion at 150 °C (T_100% = 150 °C). In contrast, the AgSmTi catalyst, which lacks Cu species, exhibits poor performance, achieving full NO_x conversion only between 225 and 250 °C. These results indicate that Ag species strongly interact with CuSmTi and play a crucial role in promoting the NH₃-SCR reaction. Figure 1b shows the N₂ selectivity of the three catalysts. The AgSmTi catalyst exhibits a sharp decrease in N₂ selectivity above 150 °C. Although the addition of Ag slightly lowers the N₂ selectivity compared to CuSmTi, the AgCuSmTi catalyst maintains over 90% selectivity across the entire temperature range of 50–300 °C. This suggests that the interaction between Ag and Cu not only enhances catalytic activity but also contributes to the preservation of high N₂ selectivity. The CO oxidation performance, shown in Figure 1c, further confirms the beneficial effect of Ag modification. Among the three catalysts, AgCuSmTi exhibits the lowest T_100% value. At any given temperature within the tested range, its CO conversion rate is slightly higher than that of CuSmTi, whereas AgSmTi shows relatively poor performance. This indicates that the synergistic interaction between Ag and CuSmTi plays a key role in enhancing CO oxidation activity as well. The TOF values for NO and CO conversion over the AgCuSmTi and CuSmTi catalysts are presented in Figure S2a and Figure S2b, respectively. The results clearly demonstrate that the incorporation of Ag significantly enhances the TOF values associated with the Cu species. Consequently, the AgCuSmTi catalyst exhibits markedly higher catalytic efficiency compared to the CuSmTi catalyst.

To further highlight the outstanding catalytic performance of the AgCuSmTi catalyst, Table S1 summarizes recent studies on Cu-based catalysts applied in NH₃-SCR coupled with CO oxidation [15,16,17,19,20,21,22,23,24]. As shown, the α-MnO₂-Cu catalyst achieves a NO_x conversion rate of 85% at 150 °C and complete CO conversion at 175 °C [25]. In comparison, the AgCuSmTi catalyst developed in this study exhibits a 100% NO_x conversion rate at the same temperature of 150 °C. Moreover, the NO conversion rate over the CuO_x/α-MnO₂ catalyst reaches only 81% at 150 °C. Several other well-reported catalysts, such as MnCuCeO_x/γ-Al₂O₃ [26], CuMgFeO [20], V₂O₅-CuO/TiO₂ [22], CuO/VWTi [11], and CuW/CeZr, exhibit higher temperature requirements for achieving maximum NO and CO conversion. The results demonstrate that the AgCuSmTi catalyst not only exhibits competitive low-temperature catalytic activity compared to traditional systems, but in fact surpasses many of them, highlighting its significant potential for practical applications in low-temperature flue gas treatment.

To evaluate the stability of the AgCuSmTi catalyst, stability tests were carried out at 150 °C and 175 °C over a period of 20 h. The results presented in Figure S3a,b clearly indicate that during the 20 h reaction duration, the conversion rates of NO and CO, as well as the N₂ selectivity, remained relatively constant. This demonstrates that the AgCuSmTi catalyst exhibits excellent stability under reaction conditions. In summary, the synergistic interaction between Ag and CuSmTi significantly enhances the catalytic performance in both the NH₃-SCR and CO oxidation reactions. The AgCuSmTi catalyst displays outstanding low-temperature activity, excellent N₂ selectivity, strong CO conversion capability, and superior stability, highlighting its significant potential for practical applications.

2.2. Structural and Chemical Characterization of the Catalyst

2.2.1. Morphological Characterization of the Catalysts

The SEM images (Figure 2a–f) reveal that both the AgCuSmTi and CuSmTi catalysts exhibit irregular nanoparticle morphologies. The incorporation of Ag effectively suppresses nanoparticle agglomeration and promotes the uniform dispersion of active species, thereby enhancing the overall catalytic performance [27]. The particle size distribution of the AgCuSmTi catalyst is presented in Figure 2g. The results indicate that the particle sizes are predominantly within the range of 14–18 nm and are relatively uniformly distributed. However, the average particle size of AgCuSmTi is 16.77 nm, which is significantly larger than that of CuSmTi (6.22 nm; see Figure S4). This suggests that the introduction of Ag has somewhat altered the particle size characteristics of the catalyst. TEM images (Figure 2h,i) show a clear lattice fringe with an interplanar spacing of approximately 0.35 nm, consistent with the (101) plane of anatase TiO₂ [28]. Notably, no distinct lattice fringes corresponding to Ag, Cu, or Sm species were observed, indicating that these metal dopants are highly dispersed on the catalyst surface rather than forming large crystalline domains or aggregates.

To further investigate the elemental composition and surface distribution characteristics of the AgCuSmTi catalyst, energy-dispersive X-ray spectroscopy (EDX) mapping analysis was conducted. As shown in Figure 2j, elements such as Ag, Cu, Sm, Ti, and O are clearly present and exhibit a uniform spatial distribution, with no signs of agglomeration. This result indicates that all metallic components are highly dispersed on the catalyst surface. This uniform elemental distribution not only confirms the successful integration and dispersion of multiple metal components but also indicates the presence of well-distributed active sites, which are crucial for efficient catalytic performance. Additionally, the high dispersion of various elements suggests that there may be strong synergistic interactions between different active components, which could help enhance catalytic activity.

2.2.2. Evidence for Ag-Cu Interaction

To further investigate the structural characteristics of the catalysts, XRD analysis was conducted, and the results are presented in Figure 3a. All samples exhibited diffraction peaks corresponding to the anatase phase of TiO₂. Notably, no characteristic diffraction peaks associated with Sm, Cu, or Ag species were observed, indicating that these elements are either amorphous or highly dispersed within the matrix [29]. Compared to pure TiO₂, the intensity of the diffraction peaks in the SmTi sample was significantly reduced, suggesting that Sm doping inhibits the growth of TiO₂ crystallites. Upon further introduction of Cu to form the CuSmTi catalyst, the peak intensities remained nearly unchanged relative to SmTi. However, upon co-doping with Ag, Cu, and Sm to obtain AgCuSmTi, the diffraction peak intensities decreased further, implying a further reduction in the average grain size of TiO₂. The smaller grain size facilitates the dispersion of active species, leading to an increased number of active sites and thereby enhancing the catalytic performance [30]. This phenomenon can be attributed to the lattice distortion and increased defect density resulting from the synergistic interaction among Ag, Cu, and Sm species, which collectively suppress the growth of TiO₂ grains.

The crystal structures of the samples were further analyzed using Raman spectroscopy. As shown in Figure 3b, all catalysts exhibit the characteristic Raman bands of anatase TiO₂ at approximately 143 cm⁻¹ (E_g), 396 cm⁻¹ (B_1g), 515 cm⁻¹ (A_1g), and 638 cm⁻¹ (E_g) [31]. These results confirm that doping does not induce any phase transformation of TiO₂ or lead to the formation of detectable secondary oxide phases, consistent with the XRD findings. However, after doping with Sm, Cu, and Ag, the Eg mode of TiO₂ undergoes noticeable blue shifts, exhibits reduced peak intensity, and shows an increased full width at half maximum (FWHM), reflecting enhanced local lattice distortions and internal stress within the TiO₂ lattice [32]. Further analysis of the changes in the E_g peak position, intensity, and FWHM reveals that the degree of lattice disturbance follows the order TiO₂ > SmTi > CuSmTi > AgCuSmTi. This trend indicates that the co-doping of Ag and Cu introduces the most significant structural perturbation. This effect is likely due to the large differences in ionic radii between Cu²⁺ (0.73 Å) and Ag⁺ (1.26 Å) compared to Ti⁴⁺ (0.64 Å), which generate considerable lattice strain when incorporated into the TiO₂ structure [33]. Moreover, Cu doping tends to create localized stress fields, while the addition of Ag further weakens the Ti-O bond strength, thereby suppressing and broadening the Raman-active vibrational modes [33,34]. In particular, the most pronounced E_g peak broadening is observed for the AgCuSmTi sample. Therefore, the combined interaction of Ag, Cu, and Sm significantly disturbs the TiO₂ lattice, resulting in blue shifting of the E_g peak, reduced intensity, and increased peak width. These structural modifications provide a fundamental basis for the enhanced catalytic performance observed in the AgCuSmTi catalyst.

Figure S5 presents the nitrogen adsorption–desorption isotherms and corresponding pore size distribution curves of the samples. As shown in Figure S5a, all catalysts display IV-type nitrogen adsorption isotherms accompanied by H₂-type hysteresis loops, indicating that they possess typical mesoporous structural characteristics. The pore size distributions calculated from the desorption branches further support this conclusion [35]. As illustrated in Figure S5b, the pore size distribution of the SmTi catalyst is only slightly affected by Cu modification. However, upon co-doping with Ag and Cu, a noticeable change in the pore size distribution is observed. The data summarized in Table S2 indicate that the incorporation of Ag and Cu leads to an increase in average pore size and a broader pore size distribution. This suggests that the synergistic interaction among Ag, Cu, and Sm alters the aggregation behavior of the particles, thereby influencing both the pore structure and specific surface area. Moreover, by comparing the catalytic performance with the specific surface area data, it can be inferred that the specific surface area is not the dominant factor affecting catalytic performance.

To further investigate the chemical states of the elements in the AgCuSmTi catalyst and its reference samples, XPS analysis was carried out. As shown in Figure 3c, the Ag 3d spectrum exhibits two main peaks corresponding to the Ag 3d_3/2 and Ag 3d_5/2 spin–orbit components. The binding energies at 368.3 eV and 367.6 eV are assigned to metallic Ag⁰ and oxidized Ag⁺ species in the Ag 3d_3/2 orbital, respectively [36]. According to the atomic ratio calculations presented in Table S3, the Ag⁺/(Ag⁰ + Ag⁺) ratio in the AgCuSmTi catalyst is higher than that in the AgSmTi catalyst upon Cu doping. Moreover, the binding energy of Ag⁺ in the AgCuSmTi sample shifts toward a higher value compared to that in AgSmTi. These observations suggest that the introduction of Cu promotes the oxidation of Ag⁰ to Ag⁺, indicating possible electron transfer interactions among Ag, Cu, and Sm, which may enhance the electronic affinity of Ag.

The Cu 2p spectrum is displayed in Figure 3d. The peaks at 934.2 eV and 932.5 eV correspond to Cu²⁺ and Cu⁺ species in the Cu 2p_3/2 orbital, respectively [36,37]. Compared with the CuSmTi catalyst, the proportion of Cu²⁺ in the AgCuSmTi catalyst is significantly increased, as summarized in

Table 1. Surface elemental composition of the catalyst as determined by XPS analysis.

Sample	Surface Atomic Concentration (%)					Relative Concentration Ratio (%)
	Ag	Cu	Sm	Ti	O	Cu2+/ (Cu+ + Cu2+)	Sm3+/ (Sm2+ + Sm3+)	Oβ/ (Oα + Oβ)
AgCuSmTi	8.79	2.61	5.50	24.63	58.46	82.51	60.98	77.77
CuSmTi	-	2.90	4.99	27.67	64.44	68.35	61.05	64.77

. This indicates that the incorporation of Ag improves the stability of the Cu in its respective oxidation states, further supporting the presence of synergistic electronic interactions between Ag, Cu, and Sm.

As shown in the Sm 3d spectrum (Figure 3e), all catalysts exhibit two fitted peaks at 1083.7 eV and 1081.4 eV, corresponding to Sm³⁺ and Sm²⁺ species, respectively [38]. Compared to the CuSmTi catalyst, the Sm³⁺/(Sm²⁺ + Sm³⁺) ratio in the AgCuSmTi catalyst remains nearly unchanged. This result suggests that the incorporation of Ag has little influence on the oxidation state of Sm, implying the absence of a strong direct interaction between Ag and Sm. Instead, the primary effect of Ag doping appears to be on its interaction with Cu species, rather than forming a direct synergistic relationship with Sm.

2.2.3. Structural Regulation of Oxygen

To gain deeper insight into the regulation of surface oxygen species and their role in catalytic performance, a comprehensive analysis of the oxygen structure in the catalyst was conducted. First, O 1s XPS was employed to qualitatively and quantitatively distinguish between lattice oxygen (O_latt) and surface-adsorbed oxygen (O_ads). As shown in Figure 4a, the O 1s spectrum can be deconvoluted into two characteristic peaks at 529.7–529.9 eV and 531.3–531.5 eV, corresponding to lattice oxygen (O_β) and surface-adsorbed oxygen (O_α), respectively [38]. The quantitative results summarized in Table 1 reveal that the AgCuSmTi catalyst exhibits a higher proportion of lattice oxygen compared to the CuSmTi catalyst. This enhancement may be attributed to the electronic synergy between Ag and Cu, which strengthens metal–oxygen bonding and increases the content of lattice oxygen within the crystal structure. The increased lattice oxygen content suggests that the AgCuSmTi catalyst possesses a greater reservoir of active oxygen species available during the initial reaction stage. This facilitates the rapid oxidation of reactant molecules and enhances the overall reaction rate. Concurrently, the consumption of lattice oxygen leads to the formation of oxygen vacancies, which can be readily replenished by gaseous oxygen, thereby sustaining efficient and continuous catalytic cycles [29].

Subsequently, O₂-TPD analysis was carried out on both the CuSmTi and AgCuSmTi catalysts to evaluate the mobility and desorption behavior of surface oxygen species, reflecting their potential involvement in oxidation reactions. As presented in Figure 4b, both catalysts exhibit two distinct desorption peaks: one in the low-temperature region (<300 °C), assigned to surface-adsorbed oxygen (O_α), and another in the medium-to-high temperature range (300–600 °C), corresponding to lattice oxygen (O_β) [20]. Notably, the AgCuSmTi catalyst shows a significant increase in lattice oxygen content, from 33.48% in CuSmTi to 48.25% after the introduction of Ag, indicating that Ag promotes the enrichment of lattice oxygen (Table S4). Therefore, the incorporation of Ag likely activates lattice oxygen through electronic interactions with Cu, enhancing both the reaction rate and stability of the catalyst.

To further validate changes in oxygen vacancy concentration, EPR spectroscopy was employed for sample analysis. As illustrated in Figure 4c, all catalysts display a distinct EPR signal at g = 2.003, which is characteristic of oxygen vacancies [39]. Among them, the CuSmTi catalyst exhibits the strongest signal intensity, indicating a high concentration of oxygen vacancies on its surface. In contrast, the AgCuSmTi catalyst shows a significantly weakened EPR signal, suggesting a reduction in oxygen vacancy density. It can be inferred that the interaction between Ag and Cu contributes to the suppression of oxygen defects while simultaneously facilitating the mobilization and enrichment of lattice oxygen.

In summary, the synergistic effect between Ag and Cu drives the structural regulation of oxygen species, not only creating favorable conditions for oxygen migration and release but also improving the accessibility and utilization efficiency of reactive oxygen species during the reaction. These effects collectively enhance the catalytic activity of the AgCuSmTi system in low-temperature NH₃-SCR processes.

2.2.4. Redox Properties and Surface Acidity Characterization

To further investigate the influence of active species on the catalyst surface and their reduction behavior in relation to catalytic performance, H₂-TPR characterization was carried out on all catalysts, with the results presented in Figure 5a. The reduction profile of the CuSmTi catalyst can be divided into three distinct stages: the low-temperature peak (131 °C) corresponds to the reduction of Cu²⁺ to Cu⁺; the medium-temperature peak (149 °C) is attributed to the reduction of Cu⁺ to Cu⁰; and the high-temperature peak (178 °C) is associated with the reduction of CuO clusters [40]. These findings indicate that Cu species exist in multiple oxidation states and undergo progressive reduction to the metallic state through sequential electron transfer processes. Upon Ag doping, all three reduction peaks in the AgCuSmTi catalyst shift toward lower temperatures, suggesting a notable enhancement in the reducibility of Cu species. This improvement can be attributed to the synergistic interaction between Ag and Cu, which facilitates electron transfer and promotes the activation of oxygen species. Consequently, Cu²⁺ becomes more readily reduced to Cu⁺ and ultimately to Cu⁰. Additionally, a new reduction peak appears at approximately 228 °C for the AgCuSmTi catalyst, which is assigned to the reduction of Ag₂O to metallic Ag⁰ [41]. The enhanced Ag–Cu interaction accelerates the redox reaction kinetics at lower temperatures, thereby improving the overall redox activity of the catalyst.

The acid characteristics of the catalyst surfaces were investigated using NH₃-TPD analysis for the AgCuSmTi and CuSmTi samples. The results are shown in Figure 5b, with the desorption peak areas quantitatively analyzed and summarized in

Table 2. Acidic site distribution of catalysis.

Sample	α	β	γ
AgCuSmTi	11.25%	76.86%	11.89%
CuSmTi	11.5%	68.15%	20.35%

. Both catalysts exhibit three distinct NH₃ desorption peaks: the α peak (100–150 °C) corresponds to weak Lewis acid sites; the β peak (150–300 °C) represents strong Lewis acid sites; and the γ peak (>300 °C) is associated with Brønsted acid sites [42]. Comparative analysis reveals a significant increase in the proportion of the β peak in the AgCuSmTi catalyst, from 68.15% in CuSmTi to 76.86%, indicating that Ag incorporation substantially enhances the density of strong Lewis acid sites. This facilitates NH₃ adsorption and activation, providing abundant active sites for subsequent surface reactions. In contrast, the proportion of the γ peak decreases in AgCuSmTi, reflecting a reduction in Brønsted acidity. These observations suggest that the Ag-Cu interaction plays a crucial role in modulating the surface acidic structure, favoring the formation of strong Lewis acid sites that are more conducive to the NH₃-SCR reaction mechanism [43]. As a result, the catalytic performance is significantly improved.

2.3. Proposed Reaction Mechanism Investigation

2.3.1. Proposed Reaction Mechanism of CO Oxidation

To elucidate the intermediate species and reaction mechanisms involved in CO oxidation over the AgCuSmTi composite metal oxide catalyst, in situ DRIFTS experiments were carried out under CO + O₂ co-adsorption conditions at various temperatures. The results are presented in Figure 6a. Upon exposure to CO and O₂, two main carbonate species formed via interactions between CO and surface oxygen species were observed: monodentate carbonate (1329 cm⁻¹ and 1463 cm⁻¹) [16,44] and bidentate carbonate (1042 cm⁻¹ and 1581 cm⁻¹) [45,46]. A distinct peak at 2113 cm⁻¹ was assigned to Cu⁺-CO, whose intensity decreased with increasing temperature and vanished entirely above 150 °C. A peak at 2163 cm⁻¹ was attributed to Cu²⁺-CO [47,48]. Additionally, broad bands in the range of 2285-2390 cm⁻¹ corresponded to CO₂ molecules, indicating the formation of gaseous CO₂ as a result of CO oxidation [49]. As shown in the spectra, the band at 1581 cm⁻¹, associated with bidentate carbonate, along with the peak at 1329 cm⁻¹, representing monodentate carbonate, both diminished significantly with rising temperature. However, these features remained detectable even at 300 °C. In contrast, the peaks at 1463 cm⁻¹ and 1042 cm⁻¹ exhibited an opposite trend, showing increased intensity at higher temperatures. Considering the evolution of CO₂ and Cu⁺-CO signals, it can be inferred that the thermal decomposition of carbonate species contributes significantly to the oxidation of CO at elevated temperatures.

The involvement of molecular oxygen in the CO oxidation mechanism was examined by comparing DRIFTS spectra obtained at 50 °C under conditions with and without O₂, as shown in Figure 6b. Notably, the intensity of the Cu⁺-CO peak at 2113 cm⁻¹ is significantly lower under CO + O₂ co-adsorption compared to CO-only adsorption, suggesting that O₂ partially occupies the active sites responsible for CO adsorption. Meanwhile, the spectral region between 1200 and 1700 cm⁻¹, which includes the carbonate-related peaks, remains largely unchanged regardless of O₂ presence. This indicates that O₂ does not directly influence the formation of carbonate species [15]. Notably, without molecular oxygen present, CO reacts with lattice oxygen, resulting in the production of carbonate intermediates. Based on the above observations, the CO oxidation reaction over the AgCuSmTi catalyst proceeds primarily through the Mars–van Krevelen (MvK) mechanism.

2.3.2. Proposed Reaction Mechanism of NH₃-SCR

To investigate the acid sites and intermediate species formed on the catalyst upon incorporation of Ag species, NH₃ adsorption–desorption DRIFTS measurements were carried out at various temperatures. Figure 7a presents the in situ DRIFTS spectra of NH₃ adsorption on the AgCuSmTi catalyst as a function of surface temperature from 50 to 300 °C. The broad band observed in the range of 3000–3400 cm⁻¹ is attributed to the N-H stretching of adsorbed NH₃ [50,51]. Additionally, two distinct peaks at 1668 cm⁻¹ and 1479 cm⁻¹ are assigned to NH₄⁺ species formed via NH₃ protonation on Brønsted acid sites [52], whereas the bands at 1195 cm⁻¹ and 1606 cm⁻¹ correspond to NH₃ coordinated to Lewis acid sites [50,53]. It is evident that the intensity of the Lewis acid site-related peaks is significantly higher than that of the Brønsted acid site peaks, indicating the dominant role of Lewis acidity in NH₃ adsorption on the AgCuSmTi catalyst. As the temperature increases, the intensities of these NH₃ adsorption features gradually decrease, suggesting progressive desorption and activation of NH₃. Notably, the spectral at 1308 cm⁻¹ and 1520 cm⁻¹ can be assigned to the formation of -NH₂ species generated via dehydrogenation of NH₃ [54,55]. Furthermore, as the temperature increases, a new band emerges at 1429 cm⁻¹ in the DRIFTS spectrum of the AgCuSmTi catalyst (Figure 7a), which is absent in the CuSmTi catalyst (Figure S6). This spectral feature can be attributed to the formation of -NH species on the Ag-doped catalyst surface [56]. -NH intermediates are likely involved in subsequent redox reactions with NO molecules, potentially leading to the formation of N₂O. This observation provides a plausible explanation for the reduced N₂ selectivity observed upon Ag doping.

To further explore the interaction between NO and O₂ on the catalyst surface, in situ DRIFTS analysis of NO + O₂ adsorption was also conducted under the same temperature conditions. The results are shown in Figure 7b. Several nitrate species were identified based on their characteristic vibrational frequencies. The bands at 1600 cm⁻¹ and 1243 cm⁻¹ are attributed to bridged nitrate species [57,58], while the peak at 1575 cm⁻¹ corresponds to bidentate nitrates [59]. The signals at 1439 cm⁻¹ and 1278 cm⁻¹ are associated with monodentate nitrate species [15,60]. As clearly observed in the spectra, the intensity of the monodentate nitrate peaks is significantly higher compared to that of bridged and bidentate nitrates. According to previous studies, the preferential formation of monodentate nitrates is beneficial for enhancing low-temperature NH₃-SCR performance [61,62]. This finding supports the conclusion that the AgCuSmTi catalyst exhibits excellent catalytic activity at low temperatures.

To elucidate the reaction mechanism of NH₃-SCR over the AgCuSmTi catalyst, in situ DRIFTS transient experiments were carried out. In the first set of experiments, NO + O₂ was introduced after NH₃ had been pre-adsorbed on the catalyst surface. The results are presented in Figure 7c. Following NH₃ pre-adsorption on the AgCuSmTi catalyst, characteristic peaks corresponding to Lewis acid sites (L-acid) were observed at 1177 cm⁻¹ and 1605 cm⁻¹, along with a peak at 1518 cm⁻¹ attributed to -NH₂ species and another at 1428 cm⁻¹ assigned to -NH species. After 30 min of N₂ purging, the intensities of the L-acid-related peaks slightly decreased, indicating that weakly physisorbed NH₃ was partially desorbed. Upon introduction of NO + O₂ for 1 min, slight decreases in the intensities of the L-acid and -NH₃ peaks were observed, suggesting that gaseous NO and O₂ reacted with adsorbed NH₃ species, consistent with an Eley–Rideal (E-R) reaction mechanism. At 5 min after NO + O₂ exposure, new spectral features emerged corresponding to nitrate species, bridged nitrate (1255 cm⁻¹), monodentate nitrate (1481 cm⁻¹), and bidentate nitrate (1541 cm⁻¹), confirming successful adsorption of NO on the catalyst surface [16]. As the reaction time increased, the intensities of these nitrate-related peaks continued to rise, while the peak at 1605 cm⁻¹ (attributed to L-acid sites) gradually diminished and eventually disappeared after 30 min. The reaction occurring 5 min after NO + O₂ introduction can be interpreted as the interaction between adsorbed NH₃ and adsorbed NO species, indicating that a Langmuir–Hinshelwood (L-H) mechanism also operates on the catalyst surface. In summary, the AgCuSmTi catalyst exhibits both the E-Rand L-H mechanisms during the NH₃-SCR process.

To further validate the above findings, a second set of experiments was conducted with a reversed reactant introduction order, where NH₃ was introduced after NO + O₂ had been pre-adsorbed on the catalyst. The results are shown in Figure 7d. Following 30 min of NO + O₂ pre-adsorption, characteristic peaks corresponding to bidentate nitrate (1545 cm⁻¹) and monodentate nitrate (1481 cm⁻¹ and 1267 cm⁻¹) were clearly observed. After introducing NH₃ for 1 min, a new peak at 1169 cm⁻¹ appeared, which is associated with L-acid centers. By 5 min, the intensities of the nitrate-related bands began to decrease, while signals corresponding to -NH₂ species and Brønsted acid (B-acid) centers emerged. This evolution indicates that the adsorbed NO species interacted with the incoming NH₃ via the L-H mechanism.

Figure 8 presents the proposed reaction pathway for the AgCuSmTi catalyst during the simultaneous removal of NO and CO. During the CO oxidation process, CO molecules adsorbed on the catalyst surface react with lattice oxygen to form either CO₂ or carbonate species. Upon desorption of CO₂, oxygen vacancies are generated, which are subsequently replenished by molecular oxygen, thereby facilitating continuous CO oxidation through lattice oxygen migration [63]. The detailed reaction mechanism is summarized as follows:

$$\mathrm{CO}\left(\mathrm{gas}\right)\to \mathrm{CO}\left(\mathrm{ads}\right)$$

$$\mathrm{CO}\left(\mathrm{ads}\right){+\mathrm{O}}^{2-}\left(\mathrm{latt}\right)\to {\mathrm{CO}}_2\left(\mathrm{g}\right)+\square \mathrm{v}$$

$$\mathrm{CO}\left(\mathrm{ads}\right){+2\mathrm{O}}^{2-}\left(\mathrm{latt}\right)\to {\mathrm{CO}}_3^{2-}+\square \mathrm{v}$$

$${\mathrm{CO}}_3^{2-}\to {\mathrm{CO}}_2{+\mathrm{O}}^{2-}$$

$${\mathrm{O}}_2\left(\mathrm{g}\right)+\square \mathrm{v}\to {2\mathrm{O}}^{2-}\left(\mathrm{latt}\right)$$

Here, “□v” represents an oxygen vacancy.

The NH₃-SCR reaction mechanism over the AgCuSmTi catalyst is predominantly governed by the L-H mechanism, with the E-R mechanism playing a secondary role. In this process, gaseous NH₃ is adsorbed and activated on the catalyst surface to form -NH₂ intermediates (Equations (1)–(4)). Activated lattice oxygen facilitates the generation of -NH₂ species and enhances the redox cycling during the reaction. Subsequently, the -NH₂ intermediate reacts with either gaseous NO (Equation (5)) or adsorbed nitrate species (Equations (6) and (7)) to yield N₂ and H₂O. During the reaction, the -NH₂ species can undergo further dehydrogenation to form -NH intermediates, which then react with NO species to produce N₂O (Equations (8) and (9)). The formation of N₂O reduces the overall N₂ selectivity of the catalyst.

The detailed reaction mechanism is summarized as follows:

$${\mathrm{NH}}_3\left(\mathrm{g}\right)\to {\mathrm{NH}}_3\left(\mathrm{ads}\right)$$

(1)

$${\mathrm{O}}_2\left(\mathrm{g}\right)\to {\mathrm{O}}_2\left(\mathrm{latt}\right)$$

(2)

$${\mathrm{O}}_2\left(\mathrm{latt}\right)\to {2\mathrm{O}}^{2-}$$

(3)

$${\mathrm{NH}}_3\left(\mathrm{ads}\right){+\mathrm{O}}^{2-}\to {-\mathrm{NH}}_2\left(\mathrm{ads}\right)+-\mathrm{OH}\left(\mathrm{ads}\right)$$

(4)

$${-\mathrm{NH}}_2\left(\mathrm{ads}\right)+\mathrm{NO}\left(\mathrm{g}\right)\to {\mathrm{N}}_2{+\mathrm{H}}_2\mathrm{O}$$

(5)

$$\mathrm{NO}\left(\mathrm{g}\right)\to \mathrm{NO}\left(\mathrm{ads}\right)$$

(6)

$${-\mathrm{NH}}_2\left(\mathrm{ads}\right)+\mathrm{NO}\left(\mathrm{ads}\right)\to {\mathrm{NH}}_2\mathrm{NO}\to {\mathrm{N}}_2{+\mathrm{H}}_2\mathrm{O}$$

(7)

$${-\mathrm{NH}}_2{\left(\mathrm{ads}\right)+\mathrm{O}}^{2-}\to -\mathrm{NH}\left(\mathrm{ads}\right)+-\mathrm{OH}\left(\mathrm{ads}\right)$$

(8)

$$-\mathrm{NH}\left(\mathrm{ads}\right)+\mathrm{NO}\left(\mathrm{g}\right)\to {\mathrm{N}}_2{\mathrm{O}+\mathrm{H}}^{+}$$

(9)

3. Experimental Section

3.1. Catalyst Preparation

All catalysts were synthesized using the sol–gel method. In a typical procedure, 25.2 mL of n-butanol and 2.8 mL of glacial acetic acid were first mixed uniformly, followed by the gradual addition of 2.4 mL of deionized water under continuous stirring for 30 min to obtain solution A. Subsequently, 1.88 g of Sm(NO₃)₃·6H₂O, 0.68 g of Cu(NO₃)₂·3H₂O, and a precisely weighed amount of AgNO₃ were added to solution A. The mixture was stirred continuously until complete dissolution of all components. Then, 9.6 mL of titanium tetrabutyltitanate (C₁₆H₃₆O₄Ti) was slowly introduced into the homogeneous solution, which was further stirred at room temperature for 3 h to obtain a viscous gel-like solution. The obtained sol was then aged at ambient temperature until gelation occurred. Afterward, the gel was transferred to an oven and dried at 110 °C overnight. Finally, the dried precursor was calcined in a muffle furnace at 500 °C for 4 h to obtain the Ag_xCuSmTi composite metal oxide catalysts with different Ag loadings (x = 0.5, 0.8, 1.0, 1.2, 1.5).

For comparative analysis, other catalytic systems were prepared following similar procedures. In one case, 9.6 mL of C₁₆H₃₆O₄Ti was added directly to solution A without incorporating any metal nitrates. The mixture was then dried and calcined under the same conditions to yield pure TiO₂. Additionally, three reference catalysts, SmTi, CuSmTi, and AgSmTi, were synthesized by sequentially adding the following combinations to solution A: (1) 1.88 g of Sm(NO₃)₃·6H₂O, (2) 1.88 g of Sm(NO₃)₃·6H₂O and 0.68 g of Cu(NO₃)₂·3H₂O, and (3) 1.88 g of Sm(NO₃)₃·6H₂O and a certain amount of AgNO₃. Each mixture was processed through the same drying and calcination steps as described above.

3.2. Catalyst Activity Test

The catalytic performance was evaluated using a fixed-bed reactor system. Detailed descriptions of the experimental setup, operating procedures, test conditions, and conversion rate calculations are provided in the Supporting Information.

3.3. Reagents and Gases

The reagents used in the preparation of the catalysts and the gases employed in the evaluation of their catalytic performance are listed in the Supporting Information.

3.4. Catalyst Characterization

The physical properties of the catalyst were investigated using a range of characterization techniques, including X-ray diffraction (XRD), N₂ adsorption, scanning electron microscopy (SEM), transmission electron microscopy (TEM), and Raman spectroscopy. Furthermore, the redox properties and acidic characteristics of the catalyst were systematically analyzed using X-ray photoelectron spectroscopy (XPS), H₂ temperature-programmed reduction (H₂-TPR), O₂ temperature-programmed desorption (O₂-TPD), NH₃ temperature-programmed desorption (NH₃-TPD), electron paramagnetic resonance (EPR), and in situ DRIFTS. Additional experimental details are provided in the Supporting Information.

4. Conclusions

This study presents an innovative approach in which Ag is introduced to modify the CuSmTi catalyst, resulting in the successful development of a dual-functional AgCuSmTi catalyst capable of effectively regulating oxygen structure. A combination of characterization techniques, including XRD, N₂ adsorption, and XPS, revealed a pronounced synergistic interaction between Ag and Cu, which plays a critical role in enhancing the catalyst’s low-temperature activity. Firstly, the Ag-Cu synergy significantly improves the physical properties of the catalyst by optimizing the dispersion of active sites, thereby establishing a structural foundation for its superior performance. Secondly, this synergistic effect markedly enhances the mobility of lattice oxygen. The highly active lattice oxygen not only facilitates redox cycling during the reaction but also accelerates electron transfer between metal species. Furthermore, the Ag-Cu interaction effectively modulates the acidic properties of the catalyst surface, increasing the density of Lewis acid sites, which are particularly favorable for the NH₃-SCR reaction. This enhancement promotes the adsorption and activation of reactant molecules, contributing to improved catalytic performance. These structural and chemical modifications significantly enhanced the performance of the AgCuSmTi catalyst, enabling complete conversion of NO and CO at temperatures as low as 150 °C and 175 °C, respectively, along with excellent N₂ selectivity. To further elucidate the underlying reaction mechanisms, in situ DRIFTS studies were conducted for both the NH₃-SCR and CO oxidation reactions. The results indicate that the NH₃-SCR process predominantly follows the L-H mechanism, with the E-R mechanism also playing a secondary role. Meanwhile, the CO oxidation reaction proceeds via the MvK mechanism. Overall, the results presented herein offer both a fundamental understanding and a promising theoretical foundation for the simultaneous elimination of nitrogen oxides and carbon monoxide from low-temperature industrial flue gases.