Structure–Activity Relationship Study of Mn/Fe Ratio Effects on Mn−Fe−Ce−O_x/γ-Al₂O₃ Nanocatalyst for NO Oxidation and Fast SCR Reaction

Abstract

A series of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts were synthesized with different Mn/Fe ratios for the catalytic oxidation of NO into NO₂ and the catalytic elimination of NO_x via fast selective catalytic reduction (SCR) reaction. The effects of Mn/Fe ratio on the physicochemical properties of the samples were analyzed by means of various techniques including N₂ adsorption, scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), H₂-temperature-programmed reduction (TPR), NH₃-temperature-programmed desorption (TPD) and NO-TPD, meanwhile, their catalytic performance was also evaluated and compared. Multiple characterizations revealed that the catalytic performance was highly dependent on the phase composition. The Mn15Fe15−Ce/Al sample with the Mn/Fe molar ratio of 1.0 presented the optimal structure characteristic among all tested samples, with the largest surface area, increased active components distributions, the reduced crystallinity and diminished particle sizes. In the meantime, the ratios of Mn⁴⁺/Mnⁿ⁺, Fe²⁺/Feⁿ⁺ and Ce³⁺/Ceⁿ⁺ in Mn15Fe15−Ce/Al samples were improved, which could enhance the redox capacity and increase the quantity of chemisorbed oxygen and oxygen vacancy, thus facilitating NO oxidation into NO₂ and eventually promoting the fast SCR reaction. In accord with the structure results, the Mn15Fe15−Ce/Al sample exhibited the highest NO oxidation rate of 64.2% at 350 °C and the broadest temperature window of 75–350 °C with the NO_x conversion >90%. Based on the structure–activity relationship discussion, the catalytic mechanism over the Mn−Fe−Ce ternary components supported by γ-Al₂O₃ were proposed. Overall, it was believed that the optimization of Mn/Fe ratio in Mn−Fe−Ce/Al nanocatalyst was an extremely effective method to improve the structure–activity relationships for NO pre-oxidation and the fast SCR reaction.

1. Introduction

Nitrogen oxides (NO_x) are one of the strong contributing factors of air pollutants, which result in acid rain, global warming and ozone depletion via photochemical redox [1]. The selective catalytic reduction (SCR) of NO_x by NH₃ or urea over high efficient catalysts is the most economical and effective technology to eliminate the pollution of NO_x [2,3]. The main SCR reaction involves the typical standard SCR reaction (1), the fast SCR reaction (2) and the NO₂ SCR reaction (3) [4,5]:

$${4\mathrm{NH}}_3+4\mathrm{NO}+{\mathrm{O}}_2\to {4\mathrm{N}}_2+{6\mathrm{H}}_2\mathrm{O}$$

(1)

$${2\mathrm{NH}}_3+\mathrm{NO}+{\mathrm{NO}}_2\to {2\mathrm{N}}_2+{3\mathrm{H}}_2\mathrm{O}$$

(2)

$${4\mathrm{NH}}_3+{3\mathrm{NO}}_2\to 3.5{\mathrm{N}}_2+{6\mathrm{H}}_2\mathrm{O}$$

(3)

The catalytic mechanism of the main SCR reaction has been systematically researched during the recent years, and the reaction processes of fast SCR have also been proposed in detail [4]. Koebel et al. [6] revealed the fast SCR reaction rate was decupled larger than that of standard SCR under 200 °C. However, the typically components of NO_x were NO₂ at ~5% and NO at ~95% [2], and thus the oxidation of NO into NO₂ previous to the main SCR progress is noteworthy so as to increase the molar ratio of NO₂/NO to 1.0 for facilitating fast SCR reaction artificially.

In the SCR reaction progress, the optimized catalyst takes an important role to the NO_x removal with NH₃. The V₂O₅/TiO₂ catalyst, promoted with WO₃ or MoO₃ usually, is the most widely used commercial catalyst. The disadvantage of V₂O₅−WO₃ (MoO₃)/TiO₂ is the strict temperature window limit [7]. These vanadium-based catalysts are not efficient enough to eliminate NO_x as the catalytic temperature below 250 °C. In recent years, the catalysts suitable for the low temperature SCR have become required, and are appropriately installed in a downstream electrostatic precipitator and desulfurizer [8]. Many research groups dedicated to utilizing high efficient active elements and structure supports to optimize low-temperature SCR catalysts with super activities, excellent stabilities and wide-ranging temperature windows.

During the past few years, a large amount of low-temperature SCR catalysts making up of transition metal oxides on multifarious supports have been investigated. The Mn−Fe−O_x based catalysts, in particular, exhibited remarkable catalytic activities, such as Mn−Fe [9], Mn−Fe/TiO₂ [10] and V₂O₅−Mn−Fe attapulgite [3]. Meanwhile, it was found that Mn−Fe−O_x based catalysts could induce the oxidation reaction of NO into NO₂ during the NH₃-SCR progress [11,12]. On this basis, some researchers have recently been absorbed oxidizing NO into NO₂, due to the more high-efficiency reaction of NO₂ with NH₃ than NO [13]. Zhang et al. [13] proved the NO oxidation into NO₂ or bidentate nitrite took place on the surface of FeMnO_x/TiO₂ via the adsorbed oxygen. Fang et al. [14] revealed the high density of lattice oxygen in the noncrystalline Mn−Fe−O_x played a leading and main role in NO oxidation. But Mn−Fe−O_x based catalysts usually needed other active elements to improve their catalytic selectivity and SO₂ resistance [15]. Among various promoter, ceria performed extremely selectivity and SO₂ resistance for the low-temperature SCR in Mn−Fe−O_x based catalysts [16]. At the same time, ceria as alkaline material with excellent redox abilities to adsorb and desorb active oxygen, also exhibited promoting effects on NO oxidization into NO₂ and NO₂ absorption into nitrites or nitrates [17]. However, the effects of Fe/Mn ratio on the activity of Mn−Fe−Ce−O_x catalyst for NO oxidation and the fast SCR reaction have not been intensive researched, especially the combined effects of γ-Al₂O₃ as carriers. The γ-Al₂O₃ is an outstanding support material for NO oxidation due to the Brönsted acid sites on its surface besides its prominent surface area [18], remarkable mechanical strength [19], great thermal stability [1] and low production cost [20].

In this research, we systematically manufactured a series of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with a Mn atomic ratio (Fe balance) from 0% to 100%. Their catalytic abilities of NO oxidization and the fast SCR were researched to understand the Mn/Fe ratio effects on the reactions. The physicochemical characteristics of the nanocatalyst samples were analyzed by N₂ adsorption, scanning electron microscopy (SEM) mapping, transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), H₂-temperature-programmed reduction (H₂-TPR), NH₃-temperature-programmed desorption (NH₃-TPD) and NO-TPD, in order to reveal the structure-activity relationship. The mechanism of Mn/Fe ratio influence on Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was also discussed. The purpose of this work was to clarify the nature of Mn−Fe−Ce based materials for catalytic performance and to explore the possibility of manufacturing nanocatalysts with outstanding capabilities in both the NO oxidization into NO₂ and the fast SCR reaction progress.

2. Results and Discussion

2.1. Morphological Characterization

2.1.1. Brunauer-Emmett-Teller (BET) Measurements

In order to compare the change of physical properties of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with different Mn/Fe ratios, the test data of BET specific surface areas, pore volumes and pore diameters were summarized in

Table 1. Physical properties of the nanocatalysts with different Mn/Fe ratios.

Samples	BET Surface Area (m2/g)	Pore Volume (cm3/g)	Pore Diameter (nm)
Fe30−Ce/Al	58.2	0.33	28.76
Mn10Fe20−Ce/Al	77.4	0.56	23.84
Mn15Fe15−Ce/Al	122.7	0.73	18.06
Mn20Fe10−Ce/Al	88.2	0.54	33.60
Mn30−Ce/Al	73.3	0.50	37.99

. The specific surface area and pore volume of Fe30−Ce/Al sample (Mn(x)Fe(y)−Ce/Al for short, x and y represented the mass percentage of Mn and Fe in the nanocatalysts, respectively) were originally 58.2 m²/g and 0.33 cm³/g, respectively. As the Mn/Fe molar ratio reached 0.5, the specific surface area rose to 77.4 m²/g and the pore volume rose to 0.56 cm³/g. The maximum specific surface area of 122.7 m²/g and the maximum pore volume of 0.73 cm³/g were both achieved over Mn15Fe15−Ce/Al with the Mn/Fe molar ratio of 1.0, which was possible due to manganese addition-enhancing active components better dispersing on the nano-Al₂O₃ support [18,21]. The manganese addition into Fe−Ce−O_x/γ-Al₂O₃ promoted mesoporosity formation and at the same time suppressed the macroporosity formation, which resulted in remarkable improvements in pore volumes [12]. Further increasing the Mn/Fe molar ratio, both the specific surface area and the pore volume started declining notably.

2.1.2. Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM) and Mapping Analysis

The typical micrographs of SEM, TEM and Mappings of the as-prepared Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with different Mn/Fe ratios were presented in Figure 1. In the SEM images shown in Figure 1a–c, the Mn10Fe20−Ce/Al sample particles were irregular. There were lots of stacking particles distributing on the catalyst surface with abundant pores collapsing produced by the fast evaporation of crystal water [22]. While the Mn15Fe15−Ce/Al sample shaped unbroken mesoporosity mesh structure on the catalyst surface. The detailed morphology structures of Mn15Fe15−Ce/Al nanoparticle were further analyzed by TEM, as revealed in Figure 1d. The Mn15Fe15−Ce/Al sample exhibited fine uniform elliptic particles with a narrow size distribution and without hard aggregate. As the Mn/Fe molar ratio rising to 2.0, the Mn20Fe10−Ce/Al sample exhibited spinel microstructure with the nanoparticle size increasing notably. According to the element mappings of the Mn15Fe15−Ce/Al sample as displayed in Figure 1e–j, it was obvious that the corresponding components of manganese, iron and cerium species were highly dispersive on the catalyst surface without regional accumulations or crystallizations.

2.2. Structural Characterization

2.2.1. X-Ray Diffraction (XRD) Analysis

The XRD results of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with different Mn/Fe ratios are exhibited in Figure 2. It was obvious that the diffraction peaks in XRD patterns matched γ-Al₂O₃ with several strong and distinguished peaks at 32.3°, 37.0°, 45.3°, and 67.0° (International Centre for Diffraction Data (ICDD) PDF card # 79-1558) [23]. Although the peaks for the structure of Al₂O₃ support remained complete with different Mn/Fe molar ratios, the diffraction angles of corresponding peaks shifted towards lower angles at varying degrees. Among all tested catalysts, the Mn15Fe15−Ce/Al sample obtained the lowest diffraction angles for each peak, which manifested the potential interaction among MnO_x, FeO_x, CeO_x and Al₂O₃. For the curve of the Mn10Fe20−Ce/Al sample, the diffraction peaks coinciding with Fe₂O₃ at 2θ = 23.7°, 42.3°, 53.3°, and 63.0°, in accordance with (012), (020), (116) and (300) crystallographic plane reflections, respectively (ICDD PDF card # 88-2359) [24]. While the diffraction peaks of Fe₃O₄ or FeO were not found in the Mn10Fe20−Ce/Al sample. Comparing the curves of the Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al samples with that of Mn10Fe20−Ce/Al sample, it could be found that the diffraction peaks of Fe₂O₃ were lowered as the Mn/Fe molar ratio increased, which probably indicated the addition of manganese into Fe−Ce−O_x/γ-Al₂O₃ decreased the crystallization of Fe₂O₃. As for the Mn20Fe10−Ce/Al sample with Mn/Fe molar ratio of 2.0, the diffraction peaks coinciding with MnO_x were very complex. The characteristic peaks at 37.5°, 42.5°, 53.8°, and 64.6° could be ascribed to MnO₂ (ICDD PDF card # 89-5171), the peaks at 32.3° and 55.1° attributed to Mn₂O₃ (ICDD PDF card # 24-0508), and the peaks at 33.5°, 36.4° and 59.7° assigned to Mn₃O₄ (ICDD PDF card # 89-4837) [25]. At the same time, the peaks matched Fe₂O₃ remarkably weakened and no peaks of other FeO_x appeared.

In the curve of Mn15Fe15−Ce/Al sample, there were no evident distinct diffraction peaks of FeO_x or MnO_x were detected. These indicated the appropriate Mn/Fe molar ratio not only facilitated the dispersion of FeO_x entirely, but also enhanced the dispersion of MnO_x completely on the catalyst surface. As a result, the coexistence of iron oxides and manganese oxides increased the active species distributions, minished the particle size and reduced the crystallinity. The smaller size particles of active element species were beneficial to the fast SCR reaction [26], but the larger oxide clusters promoted NH₃ oxidation and restrained fast SCR efficiency [27]. So, the better distribution of FeO_x and MnO_x on the nanocatalyst surface had a significant impact on the fast SCR reaction. Furthermore, no evident peaks for crystalline of CeO₂ or Ce₂O₃ were observed with Mn/Fe molar ratio of 0.5~2.0 in this work, which suggested the CeO_x might be highly dispersed on the surface of Mn−Fe−Ce/Al or the crystal diameters of CeO_x were too small (<5 nm) to be distinguished.

2.2.2. X-Ray Photoelectron Spectroscopy (XPS) Analysis

In fast SCR reaction, the components of catalyst surface and the oxidation states of active species had significant effects on the catalytic activity [5]. For the purpose of exploring the atomic chemical valences and revealing the element concentrations on the catalyst surface, XPS analysis of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was undertaken. The XPS spectra of Mn 2p, Fe 2p, Ce 3d, and O 1s in the nanocatalysts were exhibited in Figure 3. The valence states of each element were confirmed numerically by Gaussian fitting, respectively. According to these fittings, the relevant binding energy and the respective atomic concentration of elements in diverse valence states on Mn−Fe−Ce−O_x/γ-Al₂O₃ surface are shown in

Table 2. Surface atomic compositions of the catalysts determined by XPS.

Samples	Binding Energy (eV)/Atomic Composition (%)
	Mn			Fe		Ce		O
	Mn2+ 2p3/2	Mn3+ 2p3/2	Mn4+ 2p3/2	Fe2+ 2p3/2	Fe3+ 2p3/2	Ce3+ 3d5/2	Ce4+ 3d5/2	Oα 1s	Oβ 1s
Mn10Fe20−Ce/Al	640.8/33.9	642.1/36.2	644.1/29.9	709.8/28.1	712.0/71.9	885.819.4	882.6/80.6	530.3/76.6	531.7/23.4
Mn15Fe15−Ce/Al	641.3/6.8	642.5/39.4	644.4/53.8	709.1/52.4	711.3/47.6	885.5/33.8	882.4/66.2	529.7/62.6	531.2/37.4
Mn20Fe10−Ce/Al	641.1/12.7	642.247.2	644.2/40.1	709.4/42.7	711.8/57.3	885.7/23.8	882.5/76.2	529.4/70.2	531.3/29.8

.

It can be seen from Figure 3a that the XPS spectra of Mn 2p in the investigated nanocatalyst samples were matched with two constituents, attributed to Mn 2p_3/2 (peak at around 642 eV) and Mn 2p_1/2 (peak at about 653 eV) existing in MnO_x simultaneously [28]. The dissymmetric peak of Mn 2p_3/2 further confirmed the complex co-existence of manganese species in various states. The spectra of Mn 2p_3/2 could be split into three peaks via the peak-fitting deconvolution. As reported in previous studies [22], the first peak at around 641.0 ± 0.3 eV associated with MnO, the second one at 642.2 ± 0.3 eV is consistent with Mn₂O₃, and the third one at 644.2 ± 0.3 eV was assigned to MnO₂, respectively. It was apparent that the three valence states of MnO_x were difficult to separate within the binding energy difference value of 3.3 eV. The surface atomic compositions and relative intensities of Mnⁿ⁺ on the catalyst surface were calculated on account of the area covered under the separated peaks, as summarized in Table 2. With the Mn/Fe molar ratio increasing from 0.5 to 1.0, the Mn²⁺ concentration on the sample surface reduced obviously from 33.9% to 6.8%; by contrast, the concentration of Mn⁴⁺ rose significantly from 29.9% to 53.8% which became larger than that of Mn³⁺. From the above results, it was proposed that the principal phase of manganese species in the Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts changed from Mn₂O₃ to MnO₂. This was also possibly ascribed to the major amount of Mn mainly existing as +4 state covering dispersedly on the catalyst surface. Keeping on increasing the Mn/Fe molar ratio to 2.0, the percentage of Mn⁴⁺ in Mnⁿ⁺ began to decline, at the same time the Mn³⁺/Mnⁿ⁺ ratio increased remarkably. The NO catalytic activity of pure manganese oxides had been ranked as MnO₂ > Mn₂O₃ > Mn₃O₄ [25]. Furthermore, there were studies revealed that the increased concentration of Mn⁴⁺ on the catalyst surface was beneficial to SCR reactions [29]. Consequently, it could be expected that the Mn15Fe15−Ce/Al nanocatalyst would provide a superior fast SCR activity compared to Mn10Fe20−Ce/Al and Mn20Fe10−Ce/Al samples.

The Fe 2p XPS spectra of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts are displayed in Figure 3b with two distinctive peaks presented for Fe 2p_3/2 (709.1–712.0 eV) and Fe 2p_1/2 (723.8–725.0 eV). The broad peak centered at 711.0 eV included two overlapped peaks, the first one at about 709.4 ± 0.4eV was related to Fe²⁺ and the second one at around 711.8 ± 0.5eV was ascribed to Fe³⁺. Meanwhile, a satellite peak of Fe³⁺ in Fe₂O₃ appeared at 718.4 eV [10]. The overlapping peaks manifested the coexistence of iron in +2 and +3 states on the catalyst surface, which were quantified in Table 2. With the Mn/Fe molar ratio increasing from 0.5 to 1.0, the percentage of Fe³⁺ in Feⁿ⁺ decreased, while the Fe²⁺ concentration was improved. These could be attributed to the synergistic effects taking place in the redox equilibrium between Fe and Mn: Fe³⁺ + Mn³⁺ ↔ Fe²⁺ + Mn⁴⁺ [30].

Figure 3c showed the Ce 3d spectra results for Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts. The Ce 3d curve contained u and v multiplets conforming to the spin orbit split 3d_5/2 and 3d_3/2 core holes, which were further split into eight peaks on the basis of binding energy, labeled as u⁰, u′, u″, u‴ and v⁰, v′, v″, v‴, respectively [31]. The two peaks labeled u′ and v′ were attributed to the 3d¹⁰4f¹ initial electronic state of Ce³⁺, and the other six bands corresponded to the 3d¹⁰4f⁰ state of Ce⁴⁺ [32]. These indicated the species of Ce³⁺ and Ce⁴⁺ coexisted on the surface of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts. As shown in Table 2, it was obvious that the Ce³⁺/Ceⁿ⁺ ratio of Mn15Fe15−Ce/Al sample was 33.8%, as calculated by Equation (4) [33], which was greater than that of the other two catalyst samples. Therefore, the Mn/Fe molar ratio of 1.0 might be positive to the conversion from Ce⁴⁺ to Ce³⁺, which resulted in an improvement of Ce³⁺ species on its surface.

$${\eta }_{{\mathrm{Ce}}^{3+}}=\frac{u^{\prime }+v^{\prime }}{u^0+u^{\prime }+u^{″}+u^{‴}+v^0+v^{\prime }+v^{″}+v^{‴}}\times 100\%$$

(4)

The generated species of Ce³⁺ were significant inducements for the formation of electric charge balance, the generation of unsaturated chemical bonds, the improvement of surface oxygen vacancies, and the development of more chemisorbed oxygen, which would be advantageous to the activate reactants adsorption or intermediate species transformations [5]. As the temperature was under 200 °C, the main catalytic reactions followed the Eley–Rideal mechanism. NH₃ absorbed on active sites to generate NH₂ and −OH with oxygen. Subsequently, the achieved NH₂ further reacted with NO forming intermediate NH₂NO, and finally the intermediate species decomposed into N₂ and H₂O [34]. Thus, it was proposed that the quantity improvement of oxygen vacancies on the catalyst was conducive to facilitating the intermediate formation and enhancing catalytic performance in the fast SCR reaction. In the crystal lattice of Mn15Fe15−Ce/Al sample, the negative charge transferred from manganese to cerium intensifying the interaction between Mn and Ce: Mn²⁺ + Ce⁴⁺ ↔ Mn³⁺ + Ce³⁺ [16,35], Mn³⁺ + Ce⁴⁺ ↔ Mn⁴⁺ + Ce³⁺ [36,37]. The course of oxygen storage and release on the Ce³⁺/Ce⁴⁺ redox couple was brought forward: Ce₂O₃ + 1/2O₂ → 2CeO₂ and 2CeO₂ → Ce₂O₃ + O* (adsorbed) [36]. Overall, it could be easier for the catalysts with more Ce³⁺ species on the surface to develop surface oxygen vacancies, which was beneficial to the oxygen adsorption to achieve chemisorbed oxygen [10].

The XPS spectra of O 1s for Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts are illustrated in Figure 3d. According to the curve-fitting results, the O 1s spectra was split into two peaks: the first peak labeled by O_α was ascribed to lattice oxygen appearing at binding energy of 529.5–530.3 eV, the second peak denoted as O_β was attributed to chemisorbed oxygen on the catalyst surface at the binding energy of 531–532 eV. Compared with the XPS spectra of the Mn10Fe20−Ce/Al sample, both binding energies of O_α and O_β in the Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al samples shifted towards lower values, which might be caused by the existence of more Mn⁴⁺ species [9]. Furthermore, the surface concentration of chemisorbed oxygen of the Mn15Fe15−Ce/Al sample reached 37.4% as shown in Table 2, which was much larger than that of the Mn10Fe20−Ce/Al and Mn20Fe10−Ce/Al samples. The chemisorbed oxygen with high mobility was regard as the most active oxygen species in the SCR process, facilitating NO oxidation to NO₂ and promoting the fast SCR reaction in the gas mixture: NO + NO₂ + 2NH₃ → 2N₂ + 3H₂O [4].

2.2.3. H₂-Temperature-Programmed Reduction (H₂-TPR) Analysis

The reduction behavior of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts were analyzed by H₂-TPR and the curves were matched by Gaussian functions as shown in Figure 4. For each of the three nanocatalyst samples, there were five main hydrogen consumption peaks locating in the range from 100 to 900 °C. Since the support of alumina had no obvious H₂ consumption peaks in the temperature range [38], all the peaks exhibited in Figure 4 could be associated with the reduction of different species of MnO_x, FeO_x and CeO_x. For Mn10Fe20−Ce/Al sample, the initial wider reduction peak (P1) from 160 °C to 440 °C was assigned to the high oxidation state of manganese ion transformation from MnO₂ reducing to Mn₂O₃ [15]. Considering the existence of Fe₂O₃ crystallization had been confirmed by XRD analysis, the second prominent reduction peak (P2) centered at 366 °C could be mainly attributed to Mn₂O₃ species combining with the majority of Fe₂O₃ to Fe₃O₄, where iron species were settled in easily reducible sites in the form of isolated ions, oligomeric clusters or nanoparticles [39]. According to a previous report [40], the Mn₂O₃ reduction consists of two processes, Mn₂O₃ to Mn₃O₄ and Mn₂O₃ to MnO. The reduction from Mn₂O₃ to Mn₃O₄ happened more easily on the initial amorphous Mn₂O₃, and the conversion from Mn₂O₃ to MnO occurred more easily at the higher temperature range. Therefore, the third reduction peak (P3) centered at 434 °C was due to the reduction process of Mn₂O₃ to MnO and the reduction of residual Fe₂O₃ to Fe₃O₄. Meanwhile, the fourth reduction peak (P4) at around 643 °C and the fifth one (P5) at approximately 761 °C were potentially associated with the reduction process of Fe₃O₄ to FeO and FeO to Fe⁰ in isolated ions, respectively [41].

In comparison, the redox characteristic of Mn15Fe15−Ce/Al sample was significantly different from that of Mn10Fe20−Ce/Al sample. The H₂-TPR curve of Mn15Fe15−Ce/Al sample exhibited as a disequilibrium bimodal pattern, and the two main reduction peaks were at around 279 °C and 353 °C, respectively. It was rather remarkable that the MnO₂ reduction peak and the Fe₂O₃ reduction peak varied drastically with the Mn/Fe molar ratio increasing from 0.5 to 1.0. As shown in Figure 4c, the increasing Mn/Fe molar ratio resulted in the significant promotion of reducibility at a low temperature range. The reduction peak P1 was much stronger and broader with the hydrogen consumption maximum obtained at a lower temperature region. Comparing the H₂-TPR curves of Mn15Fe15−Ce/Al sample with that of Mn10Fe20−Ce/Al sample, it was noteworthy that all the reduction peaks of MnO_x and FeO_x shifted towards the lower temperature regions, signifying enhanced catalytic activities at lower temperature. Similar results were achieved by Wang et al. [12] that confirmed the Mn–Fe mixtures leading to lower temperature for Fe₃O₄ reducing to FeO and FeO converting to Fe⁰ at the same time. However, as the Mn/Fe molar ratio increased to 2.0 in the Mn20Fe10−Ce/Al sample, the reduction peaks involving MnO_x and FeO_x (P1, P2 and P3) moved to higher temperature section slightly, while the reduction peaks only containing FeO_x (P4 and P5) still shifted towards lower temperature region. In consideration of the crystallization of MnO₂, Mn₂O₃ and Mn₃O₄ observed in the XRD results as shown in Figure 2, it was supposed that the actual Mn/Fe molar ratio in the MnFeO_x nano particle was reduced, and the dominant reduction peak P1 was mainly caused by the reduction of crystallographic MnO₂ [11,30].

The redox ability of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts could be determined by the starting reduction peak temperature [42], which followed the order of Mn15Fe15−Ce/Al > Mn20Fe10−Ce/Al > Mn10Fe20−Ce/Al. This indicated that the suitable Mn/Fe molar ratio enhanced the redox activities of MnO_x and FeO_x, and was further confirmed by the total H₂ consumption as shown in

Table 3. Quantitative analysis of H₂-TPR profiles.

Samples	Temperature (°C)/H2 Consumption (mmol/g)
	Peak 1	Peak 2	Peak 3	Peak 4	Peak 5	Total
Mn10Fe20−Ce/Al	292/1.45	366/2.30	434/0.80	643/0.27	761/0.11	--/4.93
Mn15Fe15−Ce/Al	278/4.86	353/0.85	432/0.16	597/0.14	742/0.10	--/6.11
Mn20Fe10−Ce/Al	289/2.92	361/2.13	433/0.51	582/0.28	723/0.13	--/5.97

. As the Mn/Fe molar ratio increased from 0.5 to 2.0, the total H₂ consumption rose from 4.93 mmol/g to 6.11 mmol/g, and then declined to 5.97 mmol/g. On account of the redox property being an important factor affecting the fast SCR reaction, it was reasonable that the Mn15Fe15−Ce/Al sample exhibited higher NO oxidation rate and fast SCR reaction activity than the Mn20Fe10−Ce/Al and Mn10Fe20−Ce/Al samples. Meanwhile, it could be noticed that there were no distinguishable reduction peaks corresponds to cerium species. According to previous research [43], the typical reduction peaks at around 450 °C and 740 °C could be associated with the surface Ce⁴⁺ converting to Ce³⁺ and the bulk Ce⁴⁺ transforming to Ce³⁺, respectively. In this work, the reduction peaks of cerium species were very weak due to the low-quality content of cerium. At the same time, the characteristic reduction peaks of cerium were overlapped with the reduction process of Mn₂O₃ to MnO and Fe₂O₃ to Fe₃O₄ at 380~500 °C (P3), and covered by the FeO to Fe⁰ reduction at 700~800 °C (P5). Figure 5 provides a graphical representation of the reduction process, in which each active component is embodied qualitatively.

Overall, on the basis of the reduction process analyzed above, the optimal Mn/Fe molar ratio for Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was 1.0 with increased H₂ consumption and lower reduction peak temperature. This indicated that the reinforced reducibility in Mn15Fe15−Ce/Al sample was potentially due to the strong interaction between MnO_x and FeO_x with the best material proportion.

2.2.4. NH₃-Temperature-Programmed Desorption (NH₃-TPD) and NO-TPD Analysis

Except for the satisfactory redox behavior, it was another pivotal factor that the acid sites on the catalyst surface were sufficient to enhance the catalytic activities in the SCR reaction [9,17]. For the purpose of intensive investigation of the complicated relationship between the surface acidity and the activities, NH₃-TPD and NO-TPD experiments were performed on Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts and the corresponding curves according to data analysis were showed in Figure 6 and Figure 7, respectively. The NH₃-TPD patterns for these catalysts were ascribed to three desorption peaks of chemisorbed NH₃ in the temperature range from 120 to 700 °C. For the Mn10Fe20−Ce/Al sample, the first weak peak (P1) was ranged at around 213 °C, which was attributed to NH₃ desorption from the weak acid sites. The second stronger peak (P2) positioned at about 383 °C approximately was ascribed to the medium-strong acid sites. The third dominating desorption peak (P3) at around 547 °C signified the distribution of strong acid sites, which exhibited a large amount of strong acid sites matching to the desorption of strongly bound ammonia on the potential Lewis acid sites [44]. Compared with the NH₃-TPD pattern of the Mn10Fe20−Ce/Al sample, the Mn15Fe15−Ce/Al sample displayed superior acidity property at the medium- and low-temperature regions, where all the three desorption peaks shifted towards lower temperatures and converted to much wider spans. The enhancements of medium-strong acid sites and weak acid sites were positive to ammonia adsorption, which was proposed that NH₃ probably adsorbed on both Lewis acid sites and Brönsted acid sites on the catalyst surface [12]. As the Mn/Fe molar ratio increased to 2.0 in the Mn20Fe10−Ce/Al sample, the NH₃ desorption peaks moved further to the lower temperature region, and meanwhile the two strong peaks of medium-strong acid sites and weak acid sites became overlapped from 150 °C to 450 °C. The desorption peak of strong acid sites apparently reduced, which would be the reason for the decrease in fast SCR activity of Mn20Fe10−Ce/Al sample.

For the purpose of accurately analyzing the total surface acidity, the desorption peaks of the Mn10Fe20−Ce/Al, Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al samples were quantitatively analyzed and are summarized in

Table 4. Quantitative analysis of NH₃-TPD profiles.

Samples	Temperature (°C)/NH3 Composition (mmol/g)				Temperature (°C)/NO Composition (mmol/g)
	Peak 1	Peak 2	Peak 3	Total	Peak 1	Peak 2	Peak 3	Total
Mn10Fe20−Ce/Al	213/0.16	383/0.32	547/0.76	--/1.24	271/0.32	438/0.56	597/0.70	--/1.58
Mn15Fe15−Ce/Al	211/0.38	381/0.68	545/0.55	--/1.61	264/0.44	433/0.67	569/0.64	--/1.75
Mn20Fe10−Ce/Al	210/0.56	371/0.71	542/0.29	--/1.56	261/0.59	423/0.69	588/0.39	--/1.67

. The total NH₃ consumption grew from 1.24 mmol/g to 1.61 mmol/g with the Mn/Fe molar ratio increasing from 0.5 to 1.0, which further confirmed the promotion effects of the Mn/Fe molar ratio on the surface acidity of nanocatalyst samples. These could be explained by the creation of more Brönsted acid sites on the catalyst surface [16]. It was noteworthy that the Mn20Fe10−Ce/Al sample exhibited the best acidity properties at the low temperature, 0.56 mmol/g at 210 °C, which was beneficial to ammonia adsorption and advantageous to the low-temperature SCR reaction. However, it was found that the blocking effects of NH₃ applied especially to the fast SCR reaction at low temperatures [45]. The inhibiting effects were probably due to a competition of NH₃ with NO for adsorption and activation onto the metal oxidizing centers on the catalyst surface, and a disadvantageous electronic interaction between the adsorbed NH₃ and the metal sites [46]. As such, the acidity property on the nanocatalyst surface was closely associated with the redox behavior. In order to obtain the optimal catalyst, it was desirable to seek an appropriate equilibrium among the surface components, the oxidation states of metal species and the acidity property.

It was obvious that the NO-TPD patterns of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts as shown in Figure 7 are quite similar to the NH₃-TPD patterns discussed above. With temperature rising from 120 to 750 °C, each of the three catalyst samples exhibited three desorption peaks of chemisorbed NO. The dominating desorption peak of Mn10Fe20−Ce/Al sample was at about 597 °C attributed to the strong Lewis acid sites. While the main desorption peaks of Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al samples were settled at around 420~430 °C, which were ascribed to the overlapped peaks of Lewis acid sites and Brönsted acid sites. With the Mn/Fe molar ratio increasing from 0.5 to 2.0, the NO desorption at low temperature apparently augmented from 0.32 mmol/g to 0.59 mmol/g, while NO desorption in the high-temperature region declined from 0.70 mmol/g to 0.39 mmol/g, which was supposed to be conducive to fast SCR at low temperature [17].

2.3. Fast SCR Activity of Mn−Fe−Ce−O_x/γ-Al₂O₃

2.3.1. Effect of Mn/Fe Ratio on NO Oxidation of NO₂

The reaction of NO oxidation into NO₂ on Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with different Mn/Fe molar ratios are exhibited in Figure 8. Pure MnO₂ or Fe₂O₃ doping on CeO_x/γ-Al₂O₃ was also studied for comparison. It was easy to see that the NO oxidation of the five tested catalyst samples improved prominently with the temperature increasing from 25 °C to 350 °C. The NO oxidation of Fe30−Ce/Al sample was only less than 5% below 150 °C. The addition of Mn species into Fe−Ce−O_x/γ-Al₂O₃ increased the NO oxidation rate obviously and expanded the lower temperature window particularly. It was obvious that the Mn15Fe15−Ce/Al sample, which comprised the best Mn/Fe molar ratio, displayed the highest oxidation ability among the tested catalysts with the NO oxidation reaching approximately 31.3% at 150 °C and 64.2% at 350 °C, respectively. As the temperature rose above 350 °C, the NO oxidation rates began to decrease significantly for all tested samples. Moreover, the Mn20Fe10−Ce/Al and Mn30−Ce/Al samples showed lower NO oxidation rate than that of Mn15Fe15−Ce/Al sample in the researched temperature window. For the Mn20Fe10−Ce/Al sample, there was a noteworthy morphological transformation on the catalyst surface, as shown in Figure 1c, and the characteristic peaks of MnO₂, Mn₂O₃ and Mn₃O₄ appeared obvious in the XRD curve as exhibited in Figure 2. Hence, it could be confirmed that the higher crystallization of MnO_x was one of the main reasons for the deterioration of NO oxidation with Mn20Fe10−Ce/Al₂O₃ catalyst.

2.3.2. Effect of Mn/Fe Ratio on fast SCR Activity

The fast SCR activity of the prepared Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with diverse Mn/Fe molar ratios at different reaction temperatures were illustrated in Figure 9. It was found that the Mn/Fe molar ratio had remarkable effects on the catalytic performances of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts. Among all analyzed catalyst samples, the Mn15Fe15−Ce/Al sample exhibited the highest fast SCR activity in the whole tested temperature window, with NO_x conversion above 90% at 75–350 °C. As illustrated in Figure 9a, the Fe30−Ce/Al sample showed the lowest catalytic activity in NO_x conversion with the temperature rising from 25 °C to 200 °C. The addition of Mn into Fe−Ce/Al sample enhanced the fast SCR activity obviously. The appropriate amount of Mn doping enhanced the catalytic activity, while the excessive Mn doping caused a negative impact. When the Mn/Fe molar ratio grew to 0.5, the NO_x conversion of Mn10Fe20−Ce/Al sample achieved approximately 90.3% at 150 °C. When the Mn/Fe molar ratio was larger than 1.0, the NO_x conversion of Mn20Fe10−Ce/Al and Mn30−Ce/Al samples began to decline slightly. However, as the temperature changed within 250–350 °C, the variation tendency of NO_x conversion with the Mn/Fe molar ratios increasing were entirely different. In the high temperature range, the Mn30−Ce/Al sample showed the lowest catalytic activity in NO_x conversion, and the Mn20Fe10−Ce/Al sample had as high a catalytic activity as Mn15Fe15−Ce/Al. The reduction of NO_x conversion was caused by the non-selective oxidation of NH₃ at high temperatures, which was probably due to the redox property enhancement produced by manganese addition [47]. Moreover, the NO conversion performance of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was similar to that of NO_x conversion according to Figure 9b.

The NO₂ conversion of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts could be divided into two phases as shown in Figure 9c. As the temperature below 200 °C, the NO₂ conversion of all tested catalysts improved with the temperature increasing. In this temperature range, the NO₂ conversion of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts raised initially and then reduced with the Mn/Fe molar ratio increasing. The highest NO₂ conversion was obtained over the Mn15Fe15−Ce/Al sample with the Mn/Fe molar ratio of 1.0. As the temperature above 200 °C, the NO₂ conversion of all tested catalysts stabilized at approximately 99%. Based on the Figure 9, the fast SCR activity of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts could be ranked Fe30−Ce/Al < Mn10Fe20−Ce/Al < Mn30−Ce/Al < Mn20Fe10−Ce/Al < Mn15Fe15−Ce/Al. Generally, the variation tendency of the specific surface area and the pore volume with the Mn/Fe molar ratio increasing coincided with that of fast SCR activity. Thereby, it could be believed that the fast SCR activity of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was bound up with the specific surface area as reported in prior studies [30,47].

From Figure 9d, it is obvious that the variation tendency of NH₃ conversion has a close resemblance to that of NO₂ conversion over Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts. In the range of 25–200 °C, the NH₃ conversion was slightly higher than the NO_x conversion for each Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalyst at the same temperature. Meanwhile, the NH₃ conversion reached above 99% for all tested catalysts within the temperature range of 200–350 °C. According to the results exhibited in Figure 9e, it can be basically confirmed that Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts achieved satisfying N₂ selectivities. In the test temperature region, the Mn30−Ce/Al sample presented excellent N₂ selectivity that remained stable above 97.5%. The lowest N₂ selectivity was 88.3% obtained at 350 °C over the Mn15Fe15−Ce/Al sample. In order to better understand the enhancing effect of fast SCR on NO conversion than that of standard SCR, a contrast test was carried out under NO without NO₂ as shown in Figure 9f. Comparing the curves of Figure 9b,f, it could be confirmed that the participation of NO₂ in deNO_x reations boosted the NO conversion at relatively low temperature and broadened the active temperature window. Therefore, it was regarded as an effective method to improve NO_x conversion at low temperature by oxidizing NO to NO₂ and accelerating the fast SCR reaction.

2.4. Reaction Mechanism Analysis

Based on the experiment data of physicochemical characteristics and catalytic performances of the tested catalyst samples exhibited above, the Mn/Fe molar ratio in Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts influenced the oxidation and reduction characteristics of active chemisorbed sites notably, and changed the redox activities of active components significantly.

The synergistic effect among manganese, iron and cerium increased the percent of Mn⁴⁺/Mnⁿ⁺, Fe³⁺/Feⁿ⁺ and Ce³⁺/Ceⁿ⁺, formed more lattice oxygen and plenty of oxygen vacancy on the catalyst surface [16]. Among the three tested catalyst samples in this study, the Mn15Fe15−Ce/Al sample with the Mn/Fe molar ratio of 1.0 presented the highest probability of nitrate intermediate formation via the oxidization of NO into NO₂, which was regarded as a pivotal catalytic step for accelerating the fast SCR reaction [47]. The fast SCR process mainly comprised four catalytic reactions [5]. Firstly, the disproportionation of NO₂ formed ${\mathrm{NO}}_2^{-}$ and ${\mathrm{NO}}_3^{-}$ on the catalyst [48,49]. Secondly, the formed ${\mathrm{NO}}_2^{-}$ reacted with ${\mathrm{NH}}_4^{+}$ composing the intermediate NH₄NO₂ on the catalyst surface, and then NH₄NO₂ decomposed rapidly into N₂ and H₂O. Thirdly, the formed ${\mathrm{NO}}_3^{-}$ reacted with ${\mathrm{NH}}_4^{+}$ composing the intermediate NH₄NO₃ and further reacted with NO to produce NO₂, N₂ and H₂O. At the same time, some formed ${\mathrm{NO}}_3^{-}$ interacted with NO to produce ${\mathrm{NO}}_2^{-}$ and NO₂. Among these catalytic reactions, the NO transformation to NO₂ was considered as the crucial step in the reaction mechanism of fast SCR through the nitrate process. It had been revealed that MnO₂ was abundant in active oxygen and conducive to NO oxidation to NO₂ [37]. Mn⁴⁺ had the highest catalytic activity for NO elimination in consideration of the promotion effect on the reaction of NO oxidation to NO₂, which accelerated the fast SCR reaction remarkably [30]. Meanwhile, the MnO₂ phase presented better a catalytic property than Mn₂O₃ due to its lattice structure defect [50]. Therefore, Mn⁴⁺ achieved the strongest redox ability comparing to the other valence states in MnO_x-based catalysts [28]. The increase of the Mn⁴⁺/Mnⁿ⁺ ratio in the Mn15Fe15−Ce/Al sample indicated the species transformation from Mn³⁺ and Mn²⁺ to Mn⁴⁺, and the chemical circumstance variation among the Mn10Fe20−Ce/Al, Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al samples. This variation was owing to the powerful interaction among manganese, iron and cerium with different Mn/Fe molar ratios, which had been fully proved by the results exhibited above. The increase of Mn⁴⁺ contained in the Mn15Fe15−Ce/Al sample was advantageous to NH₃ adsorption and beneficial to NO catalytic oxidation. The Mn⁴⁺=O in the catalyst could react with ${\mathrm{NH}}_4^{+}$ and NO to form Mn³⁺−OH and release N₂ and H₂O. The formed Mn³⁺−OH could interact with ${\mathrm{NO}}_3^{-}$ to turn back to Mn⁴⁺=O and generate NO₂ and H₂O. As a result, the highest Mn⁴⁺/Mnⁿ⁺ ratio made a large contribution to intensify the fast SCR reaction [30]. Therefore, the Mn15Fe15−Ce/Al sample presented superior catalytic ability than the Mn10Fe20−Ce/Al and Mn20Fe10−Ce/Al samples.

Furthermore, it is proposed that the redox couples of Fe³⁺/Fe²⁺ and Ce⁴⁺/Ce³⁺ contained in Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts strengthened the redox cycle of Mn⁴⁺/Mn²⁺ and improve the amount of active oxygen, which would further promote the catalytic reactions [9,35]. The percentage of Fe²⁺/Feⁿ⁺ and Ce³⁺/Ceⁿ⁺ improved with the Mn/Fe molar ratio of 1.0, probably attributed to the strong electron transfer between Fe²⁺ ↔ Mn⁴⁺ [50], Ce³⁺ ↔ Mn³⁺ [35] and Ce³⁺ ↔ Mn⁴⁺ [37]. Hence, it is proposed that the appropriate Mn/Fe molar ratio could develop the intimate electronic interaction among manganese, iron and cerium. Comprehensive considering the XPS, H₂-TPR, NH₃-TPD and NO-TPD analysis in this study, the Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalyst with the Mn/Fe molar ratio of 1.0 achieved the optimal physicochemical property in accordance with the catalytic performance. The possible redox catalytic pathway of fast SCR reaction over Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts is exhibited in Figure 10.

3. Materials and Methods

3.1. Catalysts’ Preparation

The Mn−Fe−Ce/Al nanocatalyst samples were prepared by the co-precipitation method. Mn(NO₃)₂ (analytical pure 50%, Sinopharm, Shanghai, China), Fe(NO₃)₃·9H₂O (analytical pure 99.9%, Sinopharm, Shanghai, China), and Ce(NO₃)₃·6H₂O (analytical pure 99.9%, Nanjing-reagent, Nanjing China) were introduced as the precursors of MnO_x, FeO_x and CeO_x, respectively. The nano γ-Al₂O₃ (ultra pure 99.99%, Aladdin, Seattle, Washington, USA) was used as the carrier for the active species. The precursors were all dissolved into deionized water followed by the addition of ammonia solution till pH rose to 8. Then, the nano γ-Al₂O₃ powder was put into the aqueous solution with continuous agitation to form a homogeneous gel. Subsequently, the gel was dried in N₂ at 150 °C for 24 h and calcinated at 450 °C for 4 h. Lastly, the generated Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts were triturated and filtered with 40–60 mesh, at which sizes the diffusion effect on the catalytic performance was less than 0.5%. Hence, it could be regarded as no diffusional control with the catalytic activity calculations in this study. Meanwhile, the same size catalyst particles were used as the object for the characterization tests. In this contribution, a series of Mn−Fe−Ce−O_x/γ-Al₂O₃ samples with different Mn/Fe ratios were prepared and all samples were made up of (Mn+Fe) 30 wt% and Ce 5 wt%. The molar ratios of Mn:Fe were calculated on the basis of the mass ratios as displayed in

Table 5. Contents of Ce, Fe and Mn on Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts.

Samples	Mn		Fe		Ce		Mn:Fe
	Mass Fraction (%)	Molar Fraction (%)	Mass Fraction (%)	Molar Fraction (%)	Mass Fraction (%)	Molar Fraction (%)	MOLAR RATIO
Fe30−Ce/Al	0	0	30.0	93.8	5.0	6.2	--
Mn10Fe20−Ce/Al	10.0	31.3	20.0	62.5	5.0	6.2	0.5:1
Mn15Fe15−Ce/Al	15.0	47.0	15.0	46.8	5.0	6.2	1:1
Mn20Fe10−Ce/Al	20.0	62.6	10.0	31.2	5.0	6.2	2:1
Mn30−Ce/Al	30.0	93.9	0	0	5.0	6.1	--

.

3.2. Catalysts’ Characterization

The Maxon Tristar II 3020 micropore-size analyzer (Micromeritics, Norcross, GA, USA) was introduced to test the N₂ adsorption isotherms of the nanocatalysts at −196 °C. The nanocatalysts were vacuum degassed at 350 °C for 10 h and then the surface areas and the pore-size distributions were measured. The specific surface areas were determined according to the BET plot linear portion. The pore-size distributions were calculated according to the desorption branch with the Barrett–Joyner–Halenda (BJH) formula. The XRD profiles were achieved by a Bruker D8 advance analyzer (Bruker, Billerica, MA, USA) with Mo K_α radiation. The diffraction intensity was tested from 10° to 90° with point counting time of 1s and step of 0.02°. International Center for Diffraction Data (ICDD) was used to distinguish the element phases in XRD patterns by comparing characteristic peaks. The surface image data on the nanocatalysts was captured by a FEI Quanta 400 FEG scanning electron microscope (FEI, Hillsboro, OR, USA), and the advanced microstructural and compositional information of the nanocatalysts was obtained by a high-resolution transmission electron microscope JEOL JEM-2010 (JEOL, Tokyo, Japan). The Micromeritics Autochem II 2920 chemical adsorption instrument (Micromeritics, Norcross, GA, USA) was employed to complete the H₂-TPR, NH₃-TPD and NO-TPD tests. In the H₂-TPR experiment, 50 mg nanocatalyst was pretreatment in He at 400 °C for 1 h, then was cooled to ambient temperature in the gas mixture of H₂ and He at 30 mL/min. The H₂ consumptions were tested within 50–850 °C at the heating rate of 10 °C/min. The experimental procedure of the NH₃-TPD and NO-TPD tests were quite similar to H₂-TPR test, with NH₃ and NO replacing H₂. XPS were analyzed via a Thermo ESCALAB 250XI under the pass energy of 46.95 eV, Al K_α radiation at 1486.6 eV, X-ray source at 150 W and binding energy precision within ± 0.3 eV. C 1s line at 284.6 eV was introduced as a reference.

3.3. Catalytic Performance Tests

The catalytic performance of Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts for NO oxidation to NO₂ and fast SCR were explored in a fixed-bed reactor comprised of a tube furnace and a temperature control unit. All gas used in this experiment was controlled by mass flowmeters. The concentrations of NO and NO₂ were incessantly documented by a German MRU MGA-5 analyzer joint with an external special detector for N₂O and NH₃, as shown in Figure 11. During the experiment of NO oxidation into NO₂, the inlet gas contained 400 ppm NO, 5% O₂, ~0.3% H₂O and N₂ as balance gas. While in the test of fast SCR catalytic activity, the inlet mixed gas included 200 ppm NO, 200 ppm NO₂, 400 ppm NH₃, 5% O₂, ~0.3% H₂O and N₂ as balance gas. In the contrast experiment, 400 ppm NO was used instead of 200 ppm NO and 200 ppm NO₂. The whole flow rate was approximately 900 mL·min⁻¹, and the mass of the nanocatalyst for each run was 8.0 g. The gas hourly space velocity (GHSV) was about 25,000 h⁻¹. The NO oxidation rate and the NO_x conversion rate were calculated by the concentrations of NO_x, NO and NO₂ according to Equations (5) and (6), where [NO_x] = [NO] + [NO₂]. The NH₃ conversion rate was calculated by the inlet and outlet NH₃ concentrations, and the N₂ selectivity was calculated by the concentrations of N₂O and NO_x, as show in Equations (7) and (8) [51,52,53]. Each experiment was repeated three times to assure the results’ accuracy.

$$\mathrm{NO}\mathrm{oxidation}\mathrm{rate}=\frac{{\left[\mathrm{NO}\right]}_{\mathrm{in}}-{\left[\mathrm{NO}\right]}_{\mathrm{out}}}{{\left[\mathrm{NO}\right]}_{\mathrm{in}}}\times 100\%$$

(5)

$${\mathrm{NO}}_x\mathrm{conversion}\mathrm{rate}=\frac{{\left[{\mathrm{NO}}_x\right]}_{\mathrm{in}}-{\left[{\mathrm{NO}}_x\right]}_{\mathrm{out}}}{{\left[{\mathrm{NO}}_x\right]}_{\mathrm{in}}}\times 100\%$$

(6)

$${\mathrm{NH}}_3\mathrm{conversion}\mathrm{rate}=\frac{{\left[{\mathrm{NH}}_3\right]}_{\mathrm{in}}-{\left[{\mathrm{NH}}_3\right]}_{\mathrm{out}}}{{\left[{\mathrm{NH}}_3\right]}_{\mathrm{in}}}\times 100\%$$

(7)

$${\mathrm{N}}_2\mathrm{selectivity}=1-\frac{2{\left[{\mathrm{N}}_2\mathrm{O}\right]}_{\mathrm{out}}}{{\left[{\mathrm{NO}}_x\right]}_{\mathrm{in}}-{\left[{\mathrm{NO}}_x\right]}_{\mathrm{out}}}\times 100\%$$

(8)

4. Conclusions

Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts with different Mn/Fe ratios were synthesized to research the catalytic oxidation of NO into NO₂ and the catalytic performance in the fast SCR reaction, in order to explore the strong interactions among manganese, iron and cerium. According to the physicochemical properties of the Mn10Fe20−Ce/Al, Mn15Fe15−Ce/Al and Mn20Fe10−Ce/Al nanocatalysts, it could be concluded that the Mn15Fe15−Ce/Al sample with Mn/Fe molar ratio of 1.0 achieved the optimal coexistence of manganese, iron and cerium, enhanced the active species distribution, and restrained the metal oxide crystallinity. In the meantime, it was found that the ratios of Mn⁴⁺/Mnⁿ⁺, Fe²⁺/Feⁿ⁺ and Ce³⁺/Ceⁿ⁺ presented the maximum in Mn15Fe15−Ce/Al sample, which were beneficial to enrich oxygen vacancy and chemisorbed oxygen on the catalyst, and were conducive to facilitate the fast SCR process by promoting the possibility of NO catalytic oxidation into NO₂. The performance of NO oxidation and fast SCR catalytic activity over Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts were in accordance with the structural results. Among all the tested samples, the Mn15Fe15−Ce/Al sample presented the highest catalytic ability in the oxidation of NO into NO₂ (64.2%, 350 °C) and the optimal NO_x conversion of fast SCR (>90%, 75~350 °C). Overall, it was believed that the optimization of the Mn/Fe ratio in Mn−Fe−Ce−O_x/γ-Al₂O₃ nanocatalysts was an extremely effective method to improve the structure–activity relationships for NO pre-oxidation and the fast SCR reaction.