Effects of SiO₂, Al₂O₃ and TiO₂ Catalyst Carriers on CO-SCR Denitration Performance of Bimetallic CuCe Catalysts

Abstract

Nitrogen oxides (NO_x) emissions pose environmental and health risks. Selective catalytic reduction (SCR) is effective for NO_x removal, and using CO as a reductant can eliminate both NO_x and CO. This study explores CuCe catalysts on SiO₂, Al₂O₃, and TiO₂ for CO-SCR. Results show catalytic activity relates to the synergy between lattice oxygen and CuCe species. TiO₂ enhances this interaction, promoting Cu⁺ and lattice oxygen for NO adsorption and dissociation. The CuCe/TiO₂ catalyst achieves 100% NO conversion at 300 °C and 40.2% at 100 °C, indicating excellent low-temperature performance. These findings are valuable for developing efficient SCR catalysts.

1. Introduction

The presence of nitrogen oxides (NO_x) as a key air pollutant in the environment not only triggers serious environmental problems such as acid rain and photochemical smog, but also has far-reaching impacts on human health [1,2,3]. The presence of both NO_x and CO in flue gasses necessitates the selective catalytic reduction (SCR) of CO. NO_x denitration by CO is therefore an advantageous approach that combines the benefits of each gas. Therefore, CO-SCR technology came into being. This technology not only effectively removes NO, but also removes CO, achieving the purpose of waste for waste. So, CO-SCR technology has become one of the most promising solutions for flue gas treatment. Nevertheless, there are still some problems in the application of CO-SCR technology, such as insufficient activity at low temperatures and poor stability.

In the development of CO-SCR catalysts, supported metal catalysts have garnered widespread attention owing to their broad applicability. Notably, Cu-Ce systems were already the subject of systematic investigations during the 1980s and 1990s [4,5,6]. However, the proper selection of active components and carriers in these catalysts is crucial for the catalytic performance. Oton et al. [7] successfully prepared a series of loaded catalysts such as Ni/Al₂O₃, Fe/Al₂O₃, Co/Al₂O₃, Pt/Al₂O₃, and NiPt/Al₂O₃ with different active components by impregnation method using Al₂O₃ as a support. The results showed that the addition of active components could significantly enhance the catalyst activity, among which the NiPt/Al₂O₃ catalyst with Pt as the active component exhibited excellent NO conversion and stability. In addition, Pan et al. [8] investigated different metal-modified Fe₂O₃ catalysts for CO-SCR. Different metal-modified Fe₂O₃ catalysts were prepared by the excess impregnation method, among which the Co-Fe₂O₃/SiO₂ catalyst exhibited good catalytic activity, and the new active sites formed by the modified metal were favorable for the CO-SCR. Meng et al. [9] successfully synthesized a hollow Mn-CeO₂@Co₃O₄ catalyst using the hard template method. The results showed that Mn doping in CeO₂@Co₃O₄ produced additional active sites and modulated the electronic structure of the catalyst. In summary, studies on the effect of oxide carriers on the performance of loaded metal catalysts for CO-SCR are still relatively limited. It is also known that the choice of carriers is extremely important, and the introduction of carriers may contribute to the distribution of catalyst active components and increase active sites.

In this study, three kinds of loaded catalysts, CuCe/SiO₂, CuCe/Al₂O₃ and CuCe/TiO₂ (CuCe exists in the form of oxides), were prepared by simple co-precipitation and impregnation method using three kinds of carriers, SiO₂, Al₂O₃ and TiO₂, and the effects of different carriers on the performance of denitrification performance of CO-SCR were investigated. Compared with reported Cu-based catalysts that typically require >350 °C for complete NO conversion under similar GHSV (

Table 1. Comparison of CO conversion and NO conversion values of different samples for CO-SCR reaction under various conditions.

Catalysts	Reaction Conditions			Temperature/°C	NO Conversion/%	CO Conversion/%	Ref.
	NO	CO	GHSV
Cu-Co/TiO2	3000 ppm	3000 ppm	10,000 h−1	250	100	76	[10]
Cu-Ce-Fe-Co/TiO2	200 ppm	200 ppm	1000 h−1	200	96	72	[10]
Cu-Fe/CNT-syn	5%	10%	60,000 h−1	500	100	50	[11]
CuO-CeO2/γ-Al2O3	5%	10%	24,000 mL·g−1·h−1	400	100	-	[12]
CuO/γ-Al2O3	5%	10%	5000 h−1	200	80	-	[13]
Cu/AlPO4	0.2 vol%	1.5 vol%	2000 mL·min−1·g−1	400	85	48	[14]
CuO/CeO2/γ-Al2O3	5%	10%	12,000 h−1	400	72	-	[15]
Cu-CeO2	500 ppm	1000 ppm	2000 cm3·h−1·g−1	225	100	70	[16]
Cu/CeO2	2400 ppm	1200 ppm	24,000 mL·h−1·g−1	240	100	45	[17]
CuCe/CNT	250 ppm	5000 ppm	1500 mL·min−1·g−1	220	96.5	-	[18]
CuCe/AC	250 ppm	5000 ppm	1500 mL·min−1·g−1	220	42.6	-	[18]
CuCe/SiC	250 ppm	5000 ppm	1500 mL·min−1·g−1	220	44.6	-	[18]
CuCe/CNT	250 ppm	5000 ppm	12,600 h−1	240	96	100	[19]
Cu/CeO2	800 ppm	1600 ppm	30,000 h−1	200	70	-	[20]
CuCe/NF	500 ppm	1000 ppm	25,000 h−1	200	100	50	[21]

), the CuCe/TiO₂ catalyst achieves 100 % NO conversion at 300 °C and maintains 40.2 % conversion at 100 °C, demonstrating competitive low-temperature activity [10,11,12,13,14,15,16,17,18,19,20,21]. This efficient catalytic performance can be attributed to the excellent properties of the TiO₂, which not only improves the dispersion of the metal, but also promotes the formation of Cu⁺ and lattice oxygen, and enhances the synergistic effect between the active components. The CO-SCR reaction over CuCe/TiO₂ proceeds via a Mars–van Krevelen mechanism centered on lattice-oxygen redox cycling. Thus, optimizing the selection of the carrier effectively enhances both the activity and stability of the catalyst. This study not only establishes a key paradigm for developing novel high-efficiency catalysts, but also demonstrates the universal value of support engineering in synergistic low-temperature flue gas denitrification and decarbonization. These findings hold immediate practical significance for upgrading ultra-low emission technologies in carbon-intensive non-power industries such as steel and cement manufacturing.

2. Results and Discussion

The TEM images displayed the microscopic structures of CuCe/SiO₂, CuCe/Al₂O₃, and CuCe/TiO₂ catalysts (Figure 1a–g). The particle size distribution of CuCe/TiO₂ (inset in Figure 1c) shows a narrow Gaussian profile centered at 17.88 ± 0.003 nm, indicating uniform dispersion of active species on the TiO₂ support. Uneven distribution of the CuCe/SiO₂ and CuCe/Al₂O₃ catalysts was observed on the surface of the carrier, characterized by the presence of large agglomerates (shaded area). In contrast, the metal oxide particles in the CuCe/TiO₂ catalysts exhibited a more uniform distribution, with no prominent irregularities evident on the TiO₂ surface [22]. The grain compositions were investigated by using High Resolution Transmission Electron Microscopy (HRTEM), and the lattices of the catalyst grains were separately observed at 5 nm resolution. In the CuCe/TiO₂ catalysts, (111) crystalline facets of TiO₂ and (111) and (200) crystalline facets of CeO₂ were observed. For the CuCe/Al₂O₃ and CuCe/SiO₂ catalysts, only (111) facets of CeO₂ were observed, and no facets of the corresponding carriers were observed, probably due to the low crystallinity of the carriers (Figure 1d,e) [12]. Lattice fringes of CeO₂ were observed in all the catalysts, and lattice fringes of CuO_x were not observed, which may be due to the low and highly dispersed content of CuO_x [23]. Moreover, in the CuCe/TiO₂ catalysts, we can observe lattice interlacing between CeO₂ and TiO₂, which may be due to the interaction between carriers and active components. An analysis was carried out on the elemental dispersion in CuCe/TiO₂ catalysts, revealing a homogeneous distribution of Cu, Ce, O, and Ti elements across the catalyst’s surface. The catalysts were also tested for Cu and Ce content, and the results are shown in

Table 2. ICP testing of all samples.

Samples	Cu wt.%	Ce wt.%
CuCe/SiO2	2.8723	12.3597
CuCe/Al2O3	2.6175	12.9358
CuCe/TiO2	2.4624	10.6724

.

The XRD spectra of the catalysts CuCe/SiO₂, CuCe/Al₂O₃ and CuCe/TiO₂ are presented in Figure 1h. The spectral data display distinctive diffraction peaks corresponding to CeO₂, SiO₂, Al₂O₃, and TiO₂, thereby suggesting that the predominant crystalline phase involved in crafting the three catalysts is CeO₂. All the three catalysts show diffraction peaks at 28.7°, 33.3°, 47.4°, 56.6°, 59.4°, 69.8°, 76.8°, 79.3°, and 88.9°, respectively, which are affiliated with the (1 1 1), (2 0 0), (2 2 0), (3 1 1), (2 2 2), (4 0 0), (3 3 1), (4 2 0) and (4 2 2) crystal planes of cubic fluorite-type CeO₂ (PDF#34-0394). In the CuCe/TiO₂ catalysts, the PDF#21-1276 standard card for TiO₂ corresponds to the Rutile phase and the PDF#21-1272 standard card corresponds to the Anatase phase. In CuCe/Al₂O₃ catalysts, the PDF#10-0425 standard card corresponds to the γ-Al₂O₃. In addition, we found a shift in the diffraction peak at 28.7°, 33.3°, 47.47° and 56.6° for CeO₂. In the CuCe/TiO₂ catalyst, it was shifted by 0.81° to a small angle, proving the entry of Cu species into the lattice of CeO₂ [24,25]. It can also be seen that the peak intensity of CeO₂ in the CuCe/TiO₂ samples is relatively weak, indicating that CeO₂ is better dispersed on the surface of the carrier TiO₂, which may be due to the stronger interaction between the metal and the carrier in the catalyst [26]. In the XRD spectra of all samples, the CuO_x crystal phase does not have very distinct peaks, which is a result of the CuO_x being in a highly dispersed state and the incorporation of CuO_x into the CeO₂ lattice. This is consistent with the TEM characterization.

Raman spectroscopy represents a promising avenue for investigating the structural and lattice defects of metal oxides, as illustrated in Figure 1i. We can observe that the peaks at 445 cm⁻¹ for all catalyst samples correspond to the octahedrally localized symmetric F_2g vibrational modes around the CeO₂ cubic fluorite structure and the oxygen vacancies, respectively [27]. Among them, the CeO₂ characteristic peaks of CuCe/Al₂O₃ and CuCe/SiO₂ catalysts are more obvious, while those of CuCe/TiO₂ catalysts are weaker, which may be due to the fact that more Cu species enter into the CeO₂ lattice under the strong interactions between the metal and the carriers, and the lattice distortion is obvious, which can be confirmed by XRD and HR-TEM as well. It is also partly due to the fact that in the CuCe/TiO₂ samples the characteristic peaks of TiO₂ are strong relative to those of CeO₂, and therefore the characteristic peaks of CeO₂ are not obvious [26,28]. The peak at 612 cm⁻¹ attributed to oxygen vacancies can be observed in the spectra of CuCe/Al₂O₃ and CuCe/SiO₂ catalysts, where the peak of CuCe/Al₂O₃ is more pronounced, which proves that the content of oxygen vacancies is higher. The higher oxygen vacancy content in the CuCe/Al₂O₃ catalyst is mainly from the carrier γ-Al₂O₃ [23]. The characteristic peaks of oxygen vacancies for the CuCe/TiO₂ catalyst were not observed because the position of the characteristic peaks of oxygen vacancies overlapped with those of TiO₂, and the peak signals of the content of TiO₂ were strong, so they were covered. From the Raman spectra of CuCe/TiO₂ catalysts, it can be observed that the characteristic peaks at 145, 197, 396, 514 and 635 cm⁻¹ are all TiO₂ characteristic peaks [29].

The N₂ adsorption–desorption isothermal curves and BJH pore size distribution curves of the catalysts are shown in Figure 1j. The N₂ adsorption–desorption isothermal curve of the CuCe/TiO₂ sample belongs to a typical type I isothermal curve, which suggests that this catalyst may have a microporous structure. The N₂ adsorption–desorption isothermal curves of CuCe/Al₂O₃ samples belonged to the typical type IV isothermal curves accompanied by the H₃ loopback hysteresis curves, indicating the presence of extensive pore structures. The H₃ hysteresis curve is usually due to the accumulation of plate-like and nanoparticle structures with narrow slit-like pores [30]. The BJH curves demonstrated that the pore-size distribution patterns of the three catalysts exhibited significant differences. The N₂ adsorption–desorption isothermal curves of the CuCe/SiO₂ samples all belonged to the typical type II isothermal curves, which indicated that the catalyst might have a macroporous structure. From the pore size distribution graph, it can be observed that the void structure of the CuCe/TiO₂ samples is uniform, and the pore size is concentrated around 15 nm, and the microporous structure is usually favorable to provide more active sites, which improves the activity of the catalyst. In addition, the microporous structure helps to increase the diffusion rate of the reactants, which may have a positive effect on the CO-SCR reaction.

XPS comprehensively characterizes the surface chemistry and oxidation states of the catalysts (Figure 2a–d). The survey spectra quantify the surface Cu atomic fractions Cu/(Cu + Ce + O + Al/Ti/Si) as 1.39%, 1.59% and 3.86% for CuCe/SiO₂, CuCe/Al₂O₃ and CuCe/TiO₂, respectively. The pronounced Cu enrichment on CuCe/TiO₂ provides a higher density of active sites. Subsequent Gaussian deconvolution of the Cu2p, Ce3d and O1s high-resolution spectra yields the detailed valence-state distributions summarized in

Table 3. Binding energy and the relative value of surface atomic concentration.

Samples	Surface Atomic Species (%)
	Cu+/Cu	Ce3+/Ce	Oα/(Oα + Oβ + Oγ)
CuCe/SiO2	24.7	15.7	7.2
CuCe/Al2O3	54.0	17.2	15.0
CuCe/TiO2	75.4	30.0	82.6

. Figure 2b displays the Cu2p spectrum. The primary peaks of Cu2p_3/2 and Cu2p_1/2 are located at 931.9 eV and 951.6 eV, respectively. The peaks at 931.6 eV and 952.3 eV correspond to Cu⁺, whereas the peaks at 933.6 eV and 953.4 eV are assigned to Cu²⁺ [12,27]. Satellite fronts at 942.3 and 961.7 eV also demonstrate the presence of Cu²⁺ and Cu⁺ in all catalysts [23,31]. XPS spectra of the CuCe/TiO₂ catalysts showed a shift in the Cu2p_3/2 peak towards a lower binding energy (0.4 eV) compared to the CuCe/SiO₂ catalysts, suggesting that the greater number of Cu⁺ can be attributed to the interaction between the active component and the TiO₂ carrier [32]. In the CuCe/TiO₂ catalyst, Cu⁺ comprises 75.4% of the surface copper, a consequence of electron transfer from TiO₂ that lowers the Cu2p binding energy. This high Cu⁺ concentration sustains the rapid redox couple Ce⁴⁺ + Cu⁺ ⇌ Ce³⁺ + Cu²⁺, amplifying Cu–Ce synergy and markedly boosting catalytic performance [33,34].

Figure 2c depicts the XPS spectra of the Ce3d in the catalyst. The spectra are segmented into v and u peaks, representing two distinct spectral clusters corresponding to the spin–orbit splitting of the Ce3d_3/2 and Ce3d_5/2 orbitals, respectively. Specifically, these are further subdivided into v, v′, v″, v‴, v‴, u, u′, u″, and u‴, where v″ and u″ are associated with Ce³⁺, while the remaining six spectral bands are assigned to Ce⁴⁺, indicating that Ce⁴⁺ is the predominant oxidation state. The presence of Ce³⁺ in small amounts has higher reactivity, and the low valence of Ce³⁺ facilitates the increase in lattice oxygen, improves oxygen mobility, and at the same time generates more unsaturated chemical bonds on the catalyst surface [35]. The simultaneous presence of Ce³⁺ and Ce⁴⁺ results in Ce⁴⁺/Ce³⁺ redox pairs on the catalyst surface. The reduction from Ce⁴⁺ to Ce³⁺ promotes the oxidation of NO to NO₂ and facilitates the storage or release of reactive oxygen species [36]. The order of Ce³⁺ content was CuCe/SiO₂ < CuCe/Al₂O₃ < CuCe/TiO₂, and the highest Ce³⁺ content was found in the CuCe/TiO₂ catalysts. This phenomenon was also attributed to the interactions generated between TiO₂ and the active components, promoting enhanced synergistic effects between Cu and Ce [37].

For the O1s spectrum presented in Figure 2d, the spectra of the catalysts underwent a fitting process resulting in the identification of three distinct peaks: O_α at 529.3 eV, O_β at 530.7 eV, and O_γ at 531.7 eV. These peaks correspond to lattice oxygen, surface hydroxyl groups (-OH), and surface adsorbed oxygen, respectively [38]. The atomic concentration of O species was computed and compiled into Table 1 using the area ratios of the respective peaks. The extremely high lattice oxygen content (82.6%) in CuCe/TiO₂ reflects the strong interfacial bonding between CuCe and TiO₂ [39,40,41].

Furthermore, EPR characterization was conducted on all catalysts to substantiate the presence of oxygen vacancies and Cu²⁺ species. As depicted in Figure 2e, a symmetric resonance peak at g = 2.003 observed universally across all samples confirms the existence of oxygen vacancies. The incorporation of oxygen vacancies has been demonstrated to augment the catalyst’s active sites, thereby enhancing its catalytic efficiency and reactivity [42]. Furthermore, oxygen vacancies have been demonstrated to influence the electronic structure and charge transfer properties of the catalyst, thereby enhancing its overall catalytic performance [43]. Figure 2f reveals distinct copper speciation through characteristic EPR signals: the peak at g = 2.07 for CuCe/SiO₂ corresponds to isolated Cu²⁺ ions, while signals at g = 2.11 for CuCe/Al₂O₃ and g = 2.14 for CuCe/TiO₂ indicate square pyramidal Cu²⁺ coordination and CuO species, respectively [44]. These differential g-values demonstrate that Cu²⁺ occupies distinct chemical environments within the three catalyst systems.

The desorption behavior of NO species was analyzed via NO-TPD. Two absorption peaks appeared for all catalyst samples in Figure 2g. The peak observed in the low-temperature range was assigned to the adsorption of species including NO, N₂O, and N₂, among others, with NO being the predominant species. The peak observed in the high-temperature region was associated with the desorption of NO₂, stemming from the decomposition of NO₃⁻ [45]. The desorption peak of the CuCe/TiO₂ catalyst in the low-temperature section (152 °C) was the most obvious, indicating that the catalyst has a strong low-temperature adsorption capacity [46]. Moreover, CuCe/TiO₂ catalyst was also the first to show a strong desorption peak in the high-temperature section (347 °C), which proved that less NO₂ was produced as a by-product, corresponding to the fact that the CuCe/TiO₂ catalyst showed a very good N₂ selectivity.

To delve deeper into the reducibility of the catalyst, H₂-TPR experiments were conducted, and the outcomes are illustrated in Figure 2h. The reduction peaks observed at lower temperatures are ascribed to copper oxide, as the reduction of pure cerium dioxide, silica, aluminum trioxide, and titanium dioxide is minimal below 400 °C. The reduction temperature of copper oxide species is around 180 °C [47]. There are six distinct reduction peaks for the CuCe/TiO₂ samples at 133 °C, 146 °C, 191 °C, 353 °C, and 456 °C. The reduction peaks at 133 °C and 146 °C correspond to the reduction of highly dispersed CuO, for the 191 °C peak the reduction corresponds to the reduction of crystalline CuO, for the 353 °C the reduction is the reduction from Cu²⁺ to Cu⁺, and for the 456 °C peak the reduction corresponds to the reduction from Ce⁴⁺ to Ce³⁺. The reduction peak at 850 °C corresponds to the reduction of Cu⁺ to Cu⁰ [48,49]. The presence of Cu⁺ in the Cu/Ce/TiO₂ catalyst is supported. In the CuCe/Al₂O₃ samples, reduction peaks were observed at 162 °C, 191 °C, 206 °C, and 586 °C. The reduction peak at 162 °C corresponds to the reduction of highly dispersed CuO, the reduction peaks at 191 °C, 206 °C correspond to the reduction of crystalline CuO, and the reduction peak at 586 °C corresponds to the reduction of Ce⁴⁺ to Ce³⁺. In the CuCe/SiO₂ samples, we observed reduction peaks at 265 °C, 485 °C. The reduction peak at 265 °C corresponds to the reduction of CuO, and the reduction peak at 485 °C corresponds to the reduction of Ce⁴⁺ to Ce³⁺ [30]. Suggests that CuCe/TiO₂ samples have a stronger reduction capacity. This may result from the TiO₂ carrier promoting synergistic interactions between CuCe species [23]. Consequently, CuCe/TiO₂ exhibits the lowest onset temperature and the highest low-temperature activity.

The CO-TPD profiles of CuCe/TiO₂, CuCe/SiO₂, and CuCe/Al₂O₃ catalysts (Figure 2i) reveal distinct desorption features reflecting support-dependent active sites. All three exhibit a peak at 200 °C attributed to weakly adsorbed CO at interfacial oxygen vacancies (O_v), with CuCe/TiO₂ showing the largest area—correlating with its superior low-temperature CO and NO conversion [40]. Catalyst-specific high-temperature behavior further elucidates support effects: the 425 °C peak for CuCe/TiO₂ signifies strongly chemisorbed CO on dispersed Cu⁺ sites stabilized by TiO₂-induced σ-π back-donation [50]; the 387 °C peak for CuCe/Al₂O₃ arises from CO on Cu²⁺-O-Al³⁺ acid-metal pairs where Al₂O₃′s strong Lewis acidity fixes Cu²⁺ [51]; while the 477 °C peak for CuCe/SiO₂ reflects CO bound to aggregated CuO particles from weak metal-support interaction on inert SiO₂. Collectively, TiO₂ enables dual-site catalysis (O_v/Cu⁺ synergy), whereas Al₂O₃ limits reducibility through Cu²⁺ stabilization, and SiO₂ promotes aggregation impeding CO activation.

The catalytic activities of CuCe/SiO₂, CuCe/Al₂O₃ and CuCe/TiO₂ catalysts were determined in a CO-SCR reaction system with airspeed GHSV = 38,000 h⁻¹. As depicted in Figure 3a,b, the NO and CO conversion rates of the three catalysts exhibit a rising trend as the temperature increases, with a notable surge in conversion rates observed between 100 and 300 °C. The difference is that the CuCe/SiO₂ and CuCe/Al₂O₃ catalysts have a higher temperature for the complete removal of NO, which is fully converted at 350 °C, while the CO conversion reaches 67.4% and 93% at this temperature point. In contrast, the CuCe/TiO₂ catalyst exhibited enhanced performance at lower temperatures, achieving a 42% efficiency in NO removal at 100 °C and complete NO conversion at 300 °C. This enhancement is attributed to the promotion of interactions between CuCe species facilitated by the introduction of TiO₂, resulting in increased Cu⁺ species and lattice oxygen, which facilitate the dissociation of NO and consequently enhance catalyst performance. In addition, the CuCe/TiO₂ catalyst enabled the almost complete conversion of CO at 300 °C, realizing the co-removal of CO + NO and reducing the pollution of exhaust gas. And the CuCe/TiO₂ catalyst showed no by-product N₂O generation in the tested temperature range, reflecting 100% N₂ selectivity (Figure 3c–e). Thermal stability assessments were conducted on the CuCe/SiO₂ (300 °C), CuCe/Al₂O₃ (300 °C), and CuCe/TiO₂ (250 °C) catalysts, as illustrated in Figure 3f. After 100 h, a slight decrease of about 3% was observed for the CuCe/SiO₂ catalyst, whereas all catalysts exhibited no noticeable decline in NO conversion throughout the 150 h stability test, demonstrating excellent catalytic stability. The results are comparable to those reported in the literature (Table 1).

To probe the reaction mechanisms of CuCe/SiO₂, CuCe/Al₂O₃, and CuCe/TiO₂ catalysts, in situ FTIR was employed to examine the adsorption behavior of CO and NO molecules individually on the catalyst surfaces. Figure 4a–c illustrates the outcomes obtained from in situ FTIR spectra analysis of CO molecules interacting with the three catalysts across a temperature range of 25–400 °C. Distinct adsorbed species were identified on the surfaces of all three catalysts. The characteristic peaks at 1215 cm⁻¹ and 1300–1600 cm⁻¹ were primarily assigned to surface-adsorbed carbonate species (υasCO₃²⁻). Notably, as the temperature increased, there was a gradual decrease in the intensity of these characteristic peaks associated with υasCO₃²⁻ [52]. Moreover, the sharp band at 2112 cm⁻¹ is unequivocally assigned to Cu⁺-CO carbonyl species. Notably, the CuCe/TiO₂ catalyst exhibits the most intense signal, which remains robust even at low temperatures which is conducive to the CO-SCR reaction [53]. This phenomenon accounts for the catalyst’s elevated catalytic activity at lower temperatures. However, it should be noted that the characteristic peaks in the range of 2300–2400 cm⁻¹ are associated with CO₂ [54,55]. Over CuCe/TiO₂ and CuCe/Al₂O₃, the CO₂ bands intensify with increasing temperature; notably, CuCe/TiO₂ exhibits the strongest signal even at low temperature, demonstrating its superior low-temperature CO oxidation efficiency and thus more effective formation of the final product. Conversely, CuCe/SiO₂ exhibits an inverted band in this region because its low-temperature activity is weak; carbonates form sluggishly upon CO introduction, leaving CO predominantly adsorbed and producing only trace amounts of CO₂ and carbonates. Consequently, the CO₂ concentration falls below the background level, giving rise to a negative absorption. Furthermore, two additional bands at 2230 and 2254 cm⁻¹ observed for the CuCe/Al₂O₃ catalyst are ascribed to weakly adsorbed Al³⁺-CO and Ce⁴⁺-CO species, corresponding to linearly coordinated CO on Lewis acid sites [56].

Figure 4d–f presents the results obtained from in situ FTIR spectra analysis of NO molecules interacting with CuCe/SiO₂, CuCe/Al₂O₃, and CuCe/TiO₂ catalysts over a temperature range of 25–400 °C. For the CuCe/SiO₂ catalysts, at 1045 cm⁻¹ and 1565 cm⁻¹ are attributed to bridged bidentate nitrates, and these above species are unstable at high temperatures and gradually decompose as the temperature increases. 1191 cm⁻¹ represents the nitrite species (υNO₂⁻) on the catalyst surface, with the characteristic peaks at 1730 cm⁻¹ attributed to the N₂O₄, and the characteristic peaks of the monodentate nitrate species on the catalyst surface are at 1295, 1325, and 1433 cm⁻¹ [57,58]. The characteristic peaks of CuCe/Al₂O₃ catalyst at 1045 cm⁻¹, 1548 cm⁻¹ are attributed to the bridged bidentate nitrate, 1160 cm⁻¹ indicates υNO₂⁻ on the catalyst surface, and the characteristic peaks of υNO₃⁻ on the catalyst surface are at 1243 cm⁻¹. While attributed to monodentate nitrate species at 1305, 1406, 1482, and 1532 cm⁻¹, and to trans-(NO)₂ at 1731 cm⁻¹ [59]. The υNO₂⁻ on the surface of the CuCe/TiO₂ catalyst, which are associated with bidentate nitrate in a bridging configuration, are observed at 1189, 1046, and 1548 cm⁻¹. The species at 1731 cm⁻¹ are attributed to bidentate nitrate in a trans-(NO)₂. The characteristic peak of the υNO₃⁻ on the catalyst surface is at 1254 cm⁻¹, while 1344, and 1435 cm⁻¹ are attributed to monodentate nitrate species [3,60,61]. Upon comparison of the characteristic peaks of NO molecular adsorption species on the surfaces of the three catalysts, it is evident that the CuCe/TiO₂ catalyst exhibits the most pronounced adsorption characteristic peaks. In addition, the CuCe/TiO₂ catalyst exhibits the most intense characteristic peak for NO adsorption in the range of 1900–1950 cm⁻¹ [62,63]. The adsorption of NO molecules on the catalyst surface may facilitate the dissociation of NO and could potentially promote the reaction of CO molecules with it.

Figure 5 presents the in situ DRIFTS spectra of CuCe/TiO₂ under a CO + NO atmosphere from 100 to 400 °C. Spectroscopic analysis reveals that the band at 1205 cm⁻¹ corresponds to b-CO₃²⁻. Its intensity diminishes with increasing temperature, confirming progressive decomposition of this intermediate. The peaks at 1341, 1411, and 1464 cm⁻¹ are assigned to m-NO₃⁻, and their synchronous decline indicates reduction of nitrates by CO to N₂ and CO₂ during heating [64,65,66]. The bridging bidentate nitrate band at 1540 cm⁻¹ follows a similar trend. The Cu⁺-CO carbonyl species at 2117 cm⁻¹ exhibits maximum intensity at 100 °C and weakens with rising temperature, reflecting continuous consumption of adsorbed CO and oxidation of Cu⁺ to Cu²⁺. The gaseous CO band at 2175 cm⁻¹ intensifies at elevated temperatures, attributed to the gradual occupation (or oxidation) of surface active sites [62,67]. The gaseous CO₂ asymmetric stretching doublet at 2343 and 2361 cm⁻¹ appears prominently at 100 °C and decreases slightly thereafter, corroborating CO₂ as the primary oxidation product. [54,68,69]. These observations collectively infer a reaction pathway wherein NO initially adsorbs and dissociates on Cu⁺/Ce³⁺ active sites to form nitrate/nitrite intermediates; subsequently, adsorbed CO reduces these intermediates to N₂ and CO₂, generating oxygen vacancies. Lattice oxygen migration then replenishes the vacancies, thereby completing the Ce³⁺⇄Ce⁴⁺ and Cu⁺ ⇄ Cu²⁺ redox cycle. Thus, the CO-SCR mechanism over CuCe/TiO₂ operates predominantly via the lattice-oxygen-mediated Mars-van Krevelen (MvK) kinetics [70].

3. Experimental

3.1. Catalysts Preparation

All chemicals of analytical grade were utilized as supplied without additional purification. We used Cu(NO₃)₂·3H₂O and Ce(NO₃)₃·6H₂O, both purchased from Sigma-Aldrich (Shanghai, China), with a purity of 99.999%. NaOH and Na₂CO₃ were obtained from Sinopharm (Shanghai, China) with 99% and 99.8% purity. In addition, we used nanoscale TiO₂, Al₂O₃ (γ-Al₂O₃) and SiO₂ from Macklin, which had particle sizes of 25 nm, 20–30 nm and 20–30 nm, and were 99.99% pure. The deionized H₂O required for the experiments was homemade.

CuCe catalysts (Cu:Ce = 1:3) with a 20% total metal content (in terms of molar ratio) were synthesized using a straightforward co-precipitation impregnation technique, and all the catalysts loaded were fabricated via identical procedures. Initially, Cu and Ce nitrates were combined in a beaker at a stoichiometric ratio of 1:3, followed by the addition of 100 mL of deionized water to formulate the solution. Subsequently, a specific quantity of TiO₂ was incorporated into the beaker and stirred for a duration of one hour. A binary alkaline co-precipitant is obtained by mixing 3 mol·L⁻¹ NaOH with 1 mol·L⁻¹ Na₂CO₃ in a 1:1 volumetric ratio; this solution is then introduced slowly and dropwise into the pre-mixed precursor solution to attain and maintain the system pH at exactly 10. Subsequently, the agitation was sustained for an extended period of 12 h to allow for overnight aging. Upon the completion of the aging process, the solution underwent filtration, followed by sequential washing until pH neutrality was achieved. The washed samples were then carefully arranged in a glass receptacle within an electric drying oven set at 80 °C for desiccation. Eventually, the dried samples were finely ground into a powdered form and subjected to roasting in a muffle furnace at 500 °C for a duration of 3 h. The catalyst prepared at this time was named CuCe/TiO₂, and the other catalysts were named CuCe/Al₂O₃ and CuCe/SiO₂. To ensure the reliability and reproducibility of the experimental results, all catalyst were prepared independently more than three times. The results of the repeated experiments showed that different batches of samples exhibited consistent catalytic performance, verifying the reproducibility of the experimental operations and the consistency of the obtained data.

3.2. Catalysts Characterization

The surface microstructure, morphological characteristics, crystalline attributes, and crystal configurations of the catalysts were examined using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The TEM analysis was conducted using the FEI Tecnai G2 F30 model (Hillsboro, OR, USA). The catalysts’ crystalline structure was evaluated and studied through X-ray single crystal diffractometry (XRD) using the Bruker D8 ADVANCE (Billerica, MA, USA) model from Germany. The experimental parameters included the utilization of Cu-Kα radiation as the X-ray source, with an operating voltage of 40 kV, a current of 40 mA, and a scanning range of 2θ = 10–90°. Characterization tests were carried out using a Raman spectrometer, which was calibrated using a silicon wafer before testing. The instrument is a Renishaw inVia (Wotton-under-Edge, Gloucestershire, UK) with a 514 nm laser, 5 mW of laser power, a measurement range of 0–2000 nm⁻¹ and 5 scans. The measurement of the catalysts’ specific surface area, pore volume, and pore size was conducted through nitrogen adsorption and desorption analysis utilizing the ASAP 2460 model nitrogen adsorption instrument from Micromeritics Instrument Corporation (Norcross, GA, USA). The experimental protocol involved the degassing of the catalysts at 100 °C under vacuum conditions, followed by cooling and subsequent N₂ adsorption–desorption measurements under liquid nitrogen. Ambient free space: 17.7267 cm³ Measured, Equilibration interval: 5 s, Sample density: 1.000 g/cm³. Determination of elemental content in catalyst by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP), instrument type: AgilemtICPOES730 (Santa Clara, CA, USA), radio frequency power: 1.0 KW, carrier gas: Ar, plasma flow: 1.5 L/min, auxiliary gas flow: 1.5 L/min, nebulizer gas flow: 0.75 L/min, Detector Mode: Axial Mode, Calibration type: linear. The catalysts underwent valence and semi-quantitative elemental analysis utilizing X-ray photoelectron spectroscopy (XPS) with the Thermo ESALAB 250XI (Waltham, MA, USA) instrument. Test conditions: radiation source Mg, voltage 120 kV, current 20 mA, energy resolution 0.5 eV (full width half height, FWHM), 3 scans, calibrated using C1s (284.8 eV). Additionally, the reduction efficiency of the catalysts was evaluated through H₂ programmed thermal reduction (H₂-TPR) using the Micromeritics ASAP 2720 instrument (Norcross, GA, USA). Utilizing the Micromeritics ASAP 2720 adsorption instrument, the catalyst underwent a series of tests. The catalyst was initially purged with nitrogen at 100 °C, then cooled and exposed to a 10% H₂/Ar mixture with a flow rate of 40 mL/min, within a test temperature range varying from 25 to 900 °C. The redox performance of the catalysts was systematically probed by temperature-programmed desorption (TPD) of NO and CO on a Quantachrome AMI-90 analyzer (Boynton Beach, FL, USA). Each catalyst was pretreated at 300 °C for 30 min under He flow, cooled to 50 °C, and then exposed to either 10% NO/He or 10% CO/He (40 mL·min⁻¹) for 60 min. After purging with He to remove physisorbed species, the temperature was ramped from 50 to 700 °C at a constant rate of 2 °C min⁻¹ while the desorbed gases were monitored by a TCD detector. Electron paramagnetic resonance (EPR) measurements were carried out at room temperature using a Bruker A300 spectrometer (Rheinstetten, Germany) to analyze the electronic structure properties of the samples. The catalyst product compositions were analyzed across various temperature settings utilizing an in situ Fourier-transform infrared (FTIR) technique with the Brooke 80 V (Karlsruhe, Germany) model instrument. The samples were pretreated under Ar atmosphere for 1 h, cooled to room temperature, and the backgrounds were collected for CO adsorption (1000 ppm), NO adsorption (500 ppm) and CO (1000 ppm) + NO (500 ppm) adsorption, respectively. Ar was used as the equilibrium gas for all the above samples with a gas flow rate of 100 mL·min⁻¹ and a warming rate of 2 °C·min⁻¹, with a spectral resolution ranging from 650–4000 cm⁻¹, and the spectra were recorded from 25–500 °C. The spectral resolution of the samples were determined by using Ar as the equilibrium gas and the gas flow rate was 100 mL·min⁻¹.

3.3. Catalytic Performance Measurement

The catalytic efficiency of the synthesized catalysts was assessed using a micro-fixed bed tubular reactor, operating with a simulated flue gas containing N₂ as the balance gas. The inlet concentrations of NO ([NO]_in) and CO ([CO]_in) were set at 500 ppm and 1000 ppm, respectively, maintaining a ratio of NO to CO at 1:2. The total gas flow rate was maintained at 100 mL·min⁻¹, with a Gas Hourly Space Velocity (GHSV) of 38,000 h⁻¹. The catalytic abilities of the catalysts were assessed using FTIR, while activity evaluations were conducted over a temperature range spanning from 25 to 500 °C, during which the NO outlet concentration ([NO]_out) was monitored. Additionally, the NO outlet concentration ([NO]_out) was measured alongside FTIR testing to evaluate the performance of the catalysts. FTIR spectra were acquired on a Thermo Scientific Nicolet iS10 spectrometer (Waltham, MA, USA) at a spectral resolution of 4 cm⁻¹. Each measurement comprised 64 scans over the 4000–650 cm⁻¹ range. Activity testing of the catalysts was conducted utilizing FTIR within the temperature range from 25 to 500 °C. The assessment involved monitoring the NO, CO, N₂O, and NO₂ outlet concentrations ([NO]_out, [CO]_out, [N₂O]_out, [NO₂]_out) to determine NO conversion, CO conversion, and N₂ selectivity. The equations for NO conversion, CO conversion, and N₂ selectivity are outlined below:

$${\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}}={\displaystyle \frac{{\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{i}\mathrm{n}}-{\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}}{{\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{i}\mathrm{n}}}}\times 100\%$$

(1)

$${\left[\mathrm{C}\mathrm{O}\right]}_{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}}={\displaystyle \frac{{\left[\mathrm{CO}\right]}_{\mathrm{in}}-{\left[\mathrm{CO}\right]}_{\mathrm{out}}}{{\left[\mathrm{CO}\right]}_{\mathrm{in}}}}\times 100\%$$

(2)

$${\left[{\mathrm{N}}_2\right]}_{\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{y}}=\left(1-{\displaystyle \frac{2{\left[{\mathrm{N}}_2\mathrm{O}\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}+{\left[{\mathrm{N}\mathrm{O}}_2\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}}{{\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{i}\mathrm{n}}-{\left[\mathrm{N}\mathrm{O}\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}}}\right)\times 100\%$$

(3)

4. Conclusions

In this investigation, three catalysts were prepared—CuCe/SiO₂, CuCe/Al₂O₃, and CuCe/TiO₂—using the co-precipitation impregnation method. The CuCe/TiO₂ catalysts showed obvious synergistic interactions among CuCe species with high lattice oxygen, more active sites, and excellent NO conversion (100 °C, 40.2%), CO conversion and stability. The existence of interaction between TiO₂ carrier and active components promoted synergistic interactions between CuCe species, which affected the valence distribution of the active component metals, generating more Cu⁺ and Ce³⁺ in favor of the increase of lattice oxygen, and help dissociate NO. The individual adsorption processes of CO and NO molecules on the surfaces of the three catalysts were examined using in situ DRIFTS, demonstrating that CuCe/TiO₂ catalysts exhibited enhanced suitability for NO removal reactions at lower temperatures. The CuCe/TiO₂ catalysts exhibited the most prominent characteristic peaks for NO adsorption in the range of 1900–1950 cm⁻¹. The increased NO adsorption on the catalyst surface not only facilitated NO dissociation on the catalyst surface but also enhanced the reaction of CO with NO. The CO-SCR reaction over CuCe/TiO₂ is driven by a Mars-van Krevelen mechanism that revolves around the redox cycling of lattice oxygen. This study offers valuable insights into the advancement of novel catalysts and improved technical approaches for enhancing industrial flue gas denitrification. However, the negative impact of O₂ remains a major obstacle for application to stationary source denitrification processes.