N₂O Direct Dissociation over Mg_xCe_yCo_1−x−yCo₂O₄ Composite Spinel Metal Oxide

Abstract

A series of Mg- and/or Ce-doped Co₃O₄ (Mg_xCo_1−xCo₂O₄ Ce_xCo_1−xCo₂O₄, Mg_xCe_yCo_1−x−yCo₂O₄) composite spinel metal-oxide catalyst was prepared by a coprecipitation method and evaluated for N₂O direct decomposition. The activity measurement results suggest that Mg_0.025Ce_0.05Co_0.925Co₂O₄ with a Mg/Ce mole ratio of 0.5 exhibited the highest N₂O conversion activity, achieving 100% N₂O conversion at T = 250 °C (35 vol % N₂O balanced by He, gas hourly space velocity (GHSV) = 30,000 h⁻¹). Characterizations using X-ray diffraction (XRD), Brunauer–Emmett–Teller method (BET), hydrogen temperature-programmed reduction (H₂-TPR), and X-ray photoelectron spectroscopy (XPS) reveal that there were three main reasons for the excellent catalytic behavior of Mg_0.025Ce_0.05Co_0.925Co₂O₄: (a) Mg and Ce co-doping could reduce the grain size of composite spinel metal oxide, which thereby significantly increased the BET specific surface area of Mg_0.025Ce_0.05Co_0.925Co₂O₄ (111.2 g·m² with respect to that of 32.5 g·m⁻² for Co₃O₄); (b) Mg and Ce co-doping could improve the redox ability of Mg_0.025Ce_0.05Co_0.925Co₂O₄, including reductions of Co³⁺ → Co²⁺ and Co²⁺ → Co⁰; and (c) Mg and Ce co-doping not only could improve the migration ability of surface atomic O, but also could increase the concentrations of surface atomic O.

1. Introduction

Nitrous oxide (N₂O) is one kind of colorless and weakly sweet gas, which was considered to be a nonpolluting gas for a long time, due to its harmlessness to human beings [1,2]. Recently, people have developed a much more profound understanding of N₂O: (a) N₂O has a global warming potential (GWP) about 310 times higher than that of CO₂, which contributes to 6% of the global greenhouse effect [3]; (b) N₂O can damage the ozone layer through a cyclic chain reaction system; (c) N₂O has a very long atmospheric lifetime, nearly 150 years, as it is hard to decompose. In light of this, N₂O emissions in some European countries have been bounded by the regulation of “Kyoto Protocol” [4]. However, the N₂O content in the atmosphere is still rising. It has been mainly released from chemical industry processes, for example, adipic acid and nitric acid industry [5]. In addition to that, with the rapid increase of the numbers of vehicles, the N₂O emitted from automobile engines also contributes to the increasing content of N₂O in the atmosphere. Therefore, N₂O emission control has become a global issue to protect our living environment.

Commonly, there are three ways for N₂O abatement, including thermal pyrolysis [6], selective catalytic reduction [7,8,9], and direct decomposition [10,11,12,13,14,15,16,17]. The advantage of thermal pyrolysis is that it does not need catalyst. However, the high treatment temperature of ~800 °C not only can cause substantial energy consumption, but can also generate large amounts of NO_x, resulting in a second air pollution problem. In addition to that, N₂O decomposition (deN₂O) efficiency of thermal pyrolysis method is low, which cannot meet increasingly strict legislations. As for the selective catalytic reduction method, its treatment temperature is much lower than that of the thermal pyrolysis method; however, further addition of reductant, hydrocarbons, and ammonia, for example, greatly increases the deN₂O expense.

As for the direct decomposition method, it can directly decompose the N₂O into N₂ and O₂ at relatively low temperatures. Moreover, it does not need any additional reductant. Therefore, the direct decomposition method serves as a candidate method for N₂O abatement. The catalysts involved in N₂O direct decomposition include a noble metal catalyst [10], metal-oxide catalyst [11,12,13,14,15], and zeolite catalyst [16,17,18]. Noble metal catalysts are expensive and can be readily contaminated by the coexistence of NO gas, which greatly hinders their practical application. Zeolite catalysts—especially Fe-ZSM-5, Fe-BEA, and Fe-FER—have been widely investigated for N₂O direct decomposition due to their excellent N₂O conversion activity; however, relatively low hydrothermal stability constitutes one of their biggest problems, especially during practical applications.

A type of metal-oxide catalyst, cobalt spinel (Co₃O₄) [11,13,14,15], has been proposed to be a promising candidate for N₂O abatement, due to its excellent deN₂O activity (achieving 100% N₂O at T = 300 °C) and high thermal stability. As reported [14], the N₂O decomposition over cobalt spinel is a classic redox reaction, involving N₂O molecule activation by electron transfer from a cobalt (II) cation to a N₂O molecule (Equation (1)). The produced oxygen species can migrate to the surface of the catalyst and finally recombine with each other, forming an O₂ molecule (Equation (2)), which closes the catalytic cycle.

N₂O + e⁻ → N₂ + O⁻

(1)

2O⁻ → O₂ + 2e⁻

(2)

According to the literature report, the specific surface area and catalytic activity can be effectively improved by doping Mg into Co₃O₄ [15]. Meanwhile, ceria (CeO₂), which is commonly used as a catalytic support, possesses good oxygen storage capacity and thermal stability, which can also effectively enhance the deN₂O catalytic activity of Co₃O₄ [14]. Therefore, in the present work, a series of single Mg- or Ce-doped Co₃O₄ (Mg_xCo_1−xCo₂O₄ or Ce_xCo_1−xCo₂O₄, respectively) and Mg, Ce co-doped Co₃O₄ (Mg_xCe_yCo_1−x−yCo₂O₄) spinel composite metal-oxide catalyst were prepared by a coprecipitation method, which was further evaluated for N₂O direct decomposition. The effects of Mg-, Ce-doping were thereafter analyzed by various characterizations using X-ray diffraction (XRD), Brunauer–Emmett–Teller method (BET), hydrogen temperature-programmed reduction (H₂-TPR), and X-ray photoelectron spectroscopy (XPS) to give detailed information on the crystal structure, grain size, specific surface area, redox ability, and surface chemical state of the prepared catalyst samples. Through the investigation of present work, a promising N₂O direct decomposition candidate with high activity is proposed.

2. Results and Discussion

2.1. XRD and BET Results

XRD was conducted for samples of Mg_0.2Co_0.8Co₂O₄, Ce_0.05Co_0.95Co₂O₄, and Mg_0.025Ce_0.05Co_0.925Co₂O₄, in order to investigate the effects of Mg-, Ce-doping on the crystal structure of Co₃O₄. The related XRD patterns are profiled in Figure 1, wherein the XRD pattern of Co₃O₄ was taken as a reference. It can be found that all the samples exhibited the characteristic diffraction peaks of the Co₃O₄ spinel phase (18.9°, 31.3°, 36.8°, 44.8°, 55.7°, 59.6°, 65.2°, JCPDS 80-1541). However, after Mg- and Ce-doping, the diffraction peaks significantly decreased, especially at 36.8°, which reveals that the grain size of Co₃O₄ decreased after Mg-, Ce-doping. As shown in

Table 1. Brunauer–Emmett–Teller (BET) specific surface area and crystal size of the catalysts.

Catalysts	BET Specific Surface Area (m2·g–1)	Crystal Size (nm)
Co3O4	32.5	22.4
Mg0.2Co0.8Co2O4	77.8	14.6
Ce0.05Co0.95Co2O4	106.5	11.8
Mg0.025Ce0.05Co0.925Co2O4	111.2	10.1

, the calculated grain sizes (Scherrer equation of Equation (3)) decreased in the order of Co₃O₄ > Mg_0.2Co_0.8Co₂O₄ > Ce_0.05Co_0.95Co₂O₄ > Mg_0.025Ce_0.05Co_0.925Co₂O₄. Similar findings were also observed by Xue et al. [13] and Stelmachowski et al. [15]. It was explained in [15] that introduction of Mg, Al into Co₃O₄ spinel could increase Co₃O₄ spinel lattice constant a, but decreased another lattice constant u (defining the position of oxygen anions in the spinel lattice), which led to final decreasing of the Co–O bond lengths in both the tetrahedral and octahedral units. Therefore, it could be deduced that the decreasing crystallization of prepared Mg-, and/or Ce-doped Co₃O₄ was due to the special interaction between doped metals (Mg, Ce), with Co located at octahedral and tetrahedral units.

The specific surface areas of Co₃O₄, Mg_0.2Co_0.8Co₂O₄, Ce_0.05Co_0.95Co₂O₄, and Mg_0.025Ce_0.05Co_0.925Co₂O₄ are also listed in Table 1. It can be found that compared with pure Co₃O₄ (32.5 m²·g⁻¹), a significant increase of the specific surface area was observed for Mg-, Ce-doped Co₃O₄ spinel metal oxides (111.2 m²·g⁻¹ for Mg_0.025Ce_0.05Co_0.925Co₂O₄). Meanwhile, the increasing order of BET surface area was consistent with the decreasing order of grain size: Co₃O₄ < Mg_0.2Co_0.8Co₂O₄ < Ce_0.05Co_0.95Co₂O₄ < Mg_0.025Ce_0.05Co_0.925Co₂O₄. This indicates that the Mg- and Ce-doping could reduce the grain size of Co₃O₄, which consequently resulted in further increases of the related specific surface area.

2.2. H₂-TPR

In this section, we describe the influence of Mg-, Ce-doping on the redox ability of Co₃O₄ (Co³⁺ → Co²⁺ and Co²⁺ → Co⁰), as studied by H₂-TPR, with the results depicted in Figure 2. Two reduction peaks, respectively labeled as PH₂-I (250–350 °C) and PH₂-II (360–480 °C), were clearly observed for all investigated samples shown in Figure 2. PH₂-I represents H₂ reduction of Co₃O₄ to CoO (Co³⁺ → Co²⁺) and PH₂-II represents H₂ reduction of CoO to Co⁰ (Co²⁺ → Co⁰). As noted, there also existed a third reduction peak for the samples of Ce_0.05Co_0.95Co₂O₄ and Mg_0.025Ce_0.05Co_0.925Co₂O₄ at T = 550 °C, which was probably related to the H₂ reduction of CeO_x [19].

In comparison with the H₂ reduction profile of Co₃O₄ (Figure 2a), a much broader reduction peak located at PH₂-II was observed for Mg_0.2Co_0.8Co₂O₄ (Figure 2b). This implies that the single Mg-doping could increase the Co²⁺ content on Mg_0.2Co_0.8Co₂O₄. It was reported that Co²⁺ oxidation by N₂O to Co³⁺ was the rate-determining step during N₂O direct dissociation over Co₃O₄ [20]. Therefore, in the present work, the increasing amount of Co²⁺ of Mg_0.2Co₂O₄ is correlated with its superior deN₂O activity with respect to that of pure Co₃O₄ (as will be described later). Different from the effect of Mg-doping, Ce-doping resulted in the reduction peak of PH₂-II being shifted to much lower temperature (Ce_0.05Co_0.95Co₂O₄ of Figure 2c). This finding implies that the single Ce-doping could increase the redox ability of Co²⁺ (Co²⁺ → Co⁰), which can also well explain the much higher deN₂O activity of Ce_0.05Co_0.95Co₂O₄ versus pure Co₃O₄. As shown in Figure 2d (Mg_0.025Ce_0.05Co_0.925Co₂O₄), when Mg and Ce were simultaneously doped into the Co₃O₄, the reduction peaks of PH₂-I and PH₂-II all moved to the lowest temperature with respect to other samples. This reveals that the co-doping of Mg and Ce on Co₃O₄ could greatly improve the redox ability of Co₃O₄ (both Co³⁺ → Co²⁺ and Co²⁺ → Co⁰). In addition to that, according to the literature reports [21,22], the reduction of Co³⁺ → Co²⁺ was correlated with the oxygen desorption process. The lower the temperature needed for Co³⁺→ Co² reduction, the easier it will be to remove the oxygen. Therefore, the H₂-TPR of Mg_0.025Ce_0.05Co_0.925Co₂O₄ could well explain why it shows the highest deN₂O activity with respect to other metal-doped Co₃O₄ catalysts.

2.3. XPS

As discussed in this section, XPS was employed to investigate the surface chemical states of abovementioned the Mg-, Ce-doped Co₃O₄ composite spinel metal oxides. The related XPS spectra of Co2p, Ce3d, and O1s are shown in Figure 3A–C, respectively. As shown in Figure 3A, the binding energies of Co2p_3/2 and Co2p_1/2 in pure Co₃O₄ were located at 779.45 eV and 794.45 eV, respectively, with a Co2p_3/2, Co2p_1/2 spin-orbital splitting energy difference of 15 eV. After the single Mg-doping (Mg_0.2Co_0.8Co₂O₄ in Figure 3A), the related Co2p_3/2 and Co 2p_1/2 shifted by 0.8 eV to a lower binding energy with respect that of Co₃O₄; however, the difference in spin-orbital splitting energy remains 15 eV. Similar findings were also found for the samples of Ce_0.05Co_0.95Co₂O₄ and Mg_0.025Ce_0.05Co_0.925Co₂O₄. The related Co2p_3/2 and Co2p_1/2 shifted to lower binding energies by 1.0 and 1.1 eV, respectively; however, the splitting energy difference remained 15 eV. In general, the decrease in the binding energy of an element means an increase in its outside nucleus electron density, which consequently results in a decreasing of the element valence state. Therefore, the band shift of Co2p to much lower binding energy for Mg-, Ce-doped Co₃O₄ samples implied that the Co²⁺ content increased after Mg-, Ce-doping. This finding was also verified by the peak analysis of Co2p_3/2, discussed later. As noted, the increasing of Co²⁺ content means increasing of the O vacancy on the Mg-, Ce-doped samples. Increasing of O vacancy could then facilitate the mobility of atomic O, which was beneficial for the catalytic reaction. Therefore, the Mg_0.025Ce_0.05Co_0.925Co₂O₄ sample, displaying the largest Co2p band shift (1.1 eV), would be much more active for N₂O decomposition, as verified by the activity measurement study.

Figure 3B displays Ce 3d_5/2 and Ce 3d_3/2 of Ce_0.05Co_0.95Co₂O₄ and Mg_0.025Ce_0.05Co_0.925Co₂O₄. It can be found that there were no obvious changes in the location of Ce3d for these two samples. This finding indicates that there is no obvious difference in Ce valence state on the surface of Ce_0.05Co_0.95Co₂O₄ and Mg_0.025Ce_0.05Co_0.925Co₂O₄.

Figure 3C shows O1s of Mg-, Ce-doped Co₃O₄ samples, wherein the O1s of Co₃O₄ was taken as a reference. It can be found that (i) after Mg-, Ce-doping, the O1s of all metal-doped Co₃O₄ composite metal-oxide catalysts shifted to lower binding energy, with respect to that of pure Co₃O₄; (ii) Mg_0.025Ce_0.05Co_0.925Co₂O₄ exhibited the largest O1s band energy shift. As reported [23,24], the decrease of O1s could reduce the interactions between Co and other metal components, which consequently enhance the catalytic activity of Co²⁺. In light of that, Mg_0.025Ce_0.05Co_0.925Co₂O₄ displaying the lowest O1s binding energy can also explain its highest deN₂O activity.

Table 2. XPS surface atomic percentage.

Sample	Surface Atomic Score Levels (%)				Co2+/Co3+
	Co	Mg	Ce	O
Co3O4	35.6	0	0	64.4	0.65
Mg0.2Co0.8Co2O4	32.7	2.3	0	65.1	0.76
Ce0.05Co0.95Co2O4	17.6	0	7.4	74.9	0.91
Mg0.025Ce0.05Co0.925Co2O4	17.5	2.4	3.7	76.5	1.11

lists the surface element content of the above-discussed Mg-, Ce-doped Co₃O₄ composite metal-oxide samples, which were measured by XPS. It can be found that after Mg-, Ce-doping, the surface content of atomic Co decreased; however, the content of atomic O increased. This indicates that addition of Mg and Ce into Co₃O₄ could increase concentration of the surface atomic O. As is well known, the surface atomic O exhibits much higher catalytic activity than that of lattice oxygen, especially for the oxidation–reduction reaction system [25]. As noted, the surface atomic O content of Mg_0.025Ce_0.05Co₂O₄ is highest (76.5%, as seen in Table 2) of all, which constitutes another main factor for its best deN₂O catalytic performance.

As stated above, the Co²⁺ content of Me-, Co-doped Co₃O₄ samples increased due to a Co2p band shift to lower values. In order to determine the Co²⁺/Co³⁺ ratio on the surface of prepared Me-, Co-doped Co₃O₄, peak analyses were conducted for Co2p_3/2, as shown in Figure 4. According to integral peak areas of Co²⁺, Co³⁺, and total surface Co content, the derived Co²⁺/Co³⁺ values were calculated and are listed in Table 2. It can be clearly seen that the Co²⁺/Co³⁺ value increased in the order of Co₃O₄ < Mg_0.2Co_0.8Co₂O₄ < Ce_0.05Co_0.95Co₂O₄ < Mg_0.025Ce_0.05Co_0.925Co₂O₄, suggesting that Mg- and Ce-doping can increased the surface content of Co²⁺, and Mg_0.025Ce_0.05Co_0.925Co₂O₄ possessed the highest Co²⁺ content with respect to other samples.

2.4. Activity Measurement

Figure 5 displays the activity measurement result of a series of Mg-doped Co₃O₄ samples (Mg_xCo_1−xCo₂O₄, x = 0.2, 0.3, 0.4, 0.5). It can be found that the deN₂O activity of Mg_xCo_1−xCo₂O₄ gradually decreased as Mg-doping increased. Among the samples, Mg_0.2Co_0.8Co₂O₄ exhibited the highest deN₂O activity, which could achieve 100% N₂O conversion at T = 450 °C. Meanwhile, compared with pure Co₃O₄ (100% N₂O conversion at T > 500 °C), much higher deN₂O activity was observed for the Mg_xCo_1−xCo₂O₄ sample, when the Mg-doping amount was below 0.4. Therefore, this activity measurement result suggests that a lower amount of Mg-doping is much more preferable.

Figure 6 depicts the activity measurement result of a series of Ce-doped Co₃O₄ samples (Ce_xCo_1−xCo₂O₄, x = 0.03, 0.05, 0.1, 0.2). The Ce-doping has an obvious promotion effect on deN₂O activity, in comparison with that of Co₃O₄. Among the samples, Ce_0.05Co_0.95Co₂O₄ exhibited the highest N₂O conversion activity, which could achieve 100% N₂O conversion at T = 350 °C. Therefore, through this activity measurement, the optimum doping amount of Ce was chosen to be 0.05.

Based on the above activity measurement results, it can be concluded that the single Mg- or Ce-doping could efficiently improve N₂O activity of the Co₃O₄ spinel catalyst, where: (i) decreasing Mg-doping content is beneficial for improving deN₂O activity; and (ii) the optimum Ce-doping content is 0.05. In light of this, a series of Mg and Ce co-doping of Co₃O₄ (Mg_xCe_yCo_1−x−yCo₂O₄) was further prepared and evaluated for N₂O direct decomposition, as shown in Figure 7. Ce-loading was chosen to be 0.05 (y = 0.05), while Mg-loading varied from 0.015 to 0.15 (x = 0.015, 0.025, 0.05, 0.1, 0.15). It can be found that, except for Mg_0.15Ce_0.05Co_0.8Co₂O₄, Mg_xCe_yCo_1−x−yCo₂O₄ samples exhibit excellent deN₂O activity, achieving T₉₀ < 300 °C. Among them, Mg_0.025Ce_0.05Co_0.925Co₂O₄, with a Mg/Ce ratio of 0.5, shows the best N₂O conversion activity, which could achieve 100% N₂O conversion at T = 250 °C. This indicates that the co-doping of Mg and Ce into Co₃O₄ could indeed improve the deN₂O activity, and the best doping content of these samples was Mg_0.025Ce_0.05Co_0.925Co₂O₄ with a Mg/Ce mole ratio of 0.5. In addition to that, the long lifetime (30 h) deN₂O activity measurement was further conducted for the best-performing Mg_0.025Ce_0.05Co_0.925Co₂O₄ sample at T = 250 °C, as shown in Figure 8. It can be found that the prepared sample exhibited stable deN₂O activity, implying excellent durability of this sample.

2.5. Correlation between Characterization and Activity Measurement Result

In order to investigate Mg-, Ce-doping effects on the catalytic performance of Co₃O₄, various characterizations, including XRD, BET, H₂-TPR, and XPS were conducted for the samples of single metal doping and Mg, Ce co-doping on Co₃O₄ (Mg_0.2Co_0.8Co₂O₄, Ce_0.05Co_0.95Co₂O₄). As revealed by XRD and BET, Mg-, Ce-doping could reduce the grain size of Co₃O₄, which in turn efficiently increases the related BET specific surface area: Co₃O₄ (32.5 m²·g⁻¹) < Mg_0.2Co_0.8Co₂O₄ (77.8 m²·g⁻¹) < Ce_0.05Co_0.95Co₂O₄ (106.5 m²·g⁻¹) < Mg_0.025Ce_0.05Co_0.925Co₂O₄ (111.2 m²·g⁻¹). Therefore, the highest specific surface of Mg_0.025Ce_0.05Co_0.925Co₂O₄ constitutes one of the important factors that led to the highest deN₂O activity. In addition to that, the improved redox abilities of Mg-, Ce-doped Co₃O₄ samples were observed based on H₂-TPR, and these also play an important role during N₂O direct dissociation: (i) single Mg-doping could increase Co²⁺ content on Mg_0.2Co_0.8Co₂O₄; (ii) single Ce-doping could improve Co²⁺ redox ability; (iii) Mg-, Ce-doping could improve both Co³⁺ and Co²⁺ redox ability; (iv) compared with other Me-, Ce-doped Co₃O₄ samples, the highest redox ability of Mg_0.2Co_0.8Co₂O₄ is another reason for its showing the highest deN₂O activity. The XPS, utilized to investigate the surface atomic component and valence state, suggested that Mg-, Ce-doping did not influence the valence status of Co. However, the O1s suggested that Mg-, Ce-doping could efficiently enhance the mobility as well as content of surface atomic O, which is favorable for N₂O dissociation. Therefore, the highest mobility as well as content of surface atomic O on Mg_0.2Co_0.8Co₂O₄ constituted another important factor of its displaying the highest deN₂O. In summary, the highest BET specific surface area, redox ability, surface atomic O mobility, as well as surface oxygen content of the prepared Mg, Ce co-doped Co₃O₄ (Mg_0.025Ce_0.05Co_0.925Co₂O₄) constitute the main reasons for this sample having the highest deN₂O activity, which also make this kind of catalyst promising for further practical applications.

3. Experimental

3.1. Preparation of Catalyst

The composite spinel metal oxide catalysts of the present work (Mg_xCo_1−xCo₂O₄, Ce_xCo_1−xCo₂O₄, and Mg_xCe_yCo_1−x−yCo₂O₄) were prepared by a coprecipitation method. Firstly, a certain amount of metal nitrate solutions (magnesium nitrate, cerium nitrate, cobalt nitrate) were mixed together and placed in a 70 °C water bath, which was stirred for 20 min. Then, an appropriate amount of Na₂CO₃ was dissolved into 20 mL deionized water in a 70 °C water bath, which was stirred for 20 min. The above two solutions were mixed and the pH value was adjusted to 9. After filtration, drying (120 °C for 2 h), and calcination (400 °C for 3 h), the final catalyst samples were obtained.

3.2. Characterization of Catalysts

The XRD patterns of the prepared samples have been recorded in 2θ ranges from 5° to 80°, based on X-ray diffractometer (Rigaku, Tokyo, Japan, D/max2500VB2) and with a radiation of Cu Kα (λ = 1.5406 Å). The crystal phases were confirmed according to the JCPDS reference. The crystallite sizes were calculated according to the Scherrer equation (Equation (3)) [3]. In this formula, K is Scherrer constant; β is the half width of the diffraction peak; and θ is a diffraction angle.

$$D=\frac{K\mathsf{\lambda }}{\mathsf{\beta }\cos \mathsf{\theta }}$$

(3)

The specific surface area (S_BET) of the prepared catalyst sample was measured via a Sorptomatic 1990 instrument (Thermo Electron, Waltham, MA, USA) through nitrogen adsorption/desorption at 77 K and was calculated by the BET method. H₂-TPR was performed using Thermo Electron TPD/R/O 1100 Series instrument equipped with a thermal conductivity detector (TCD). Before the experiment, samples were in turn pretreated by the following procedures: (1) heated at 100 °C for 1 h in a He (>99.999%) stream; (2) cooled down to 30 °C under above atmosphere; (3) H₂-TPR started with a ramp of 10 °C/min from 30 to 750 °C. The flow rate of reducing gas (5.06% H₂ balanced by N₂) was 30 mL/min. XPS was conducted on a Thermo Fisher ESCALAB 250 system (Thermo Fisher Scientific, Waltham, MA, USA) with Al Kα radiation under ultrahigh vacuum (UHV), calibrated internally by carbon deposit C (1s) having a binding energy (BE) of 284.6 eV.

3.3. Activity Measurement

N₂O direct decomposition was conducted in a fixed bed quartz tube reactor (Φ7 × 1 × L300 mm) under atmospheric pressure. Pelletized catalyst (0.2 g) with particle sizes of 250–425 μm was placed in the constant temperature zone of a vertical tubular reactor. Afterwards, a gas mixture consisting of 35 vol % N₂O balanced by 65 vol % He was made according to the components of tail gas from related industrial plants as previously described [3] and fed into the reactor under atmospheric pressure. A total flow rate of 140 mL/min was employed for the activity test, corresponding to a GHSV (gas hourly space velocity) of 30,000 h⁻¹. The gaseous products were discontinuously analyzed by gas chromatograph (GC-4000A, Dongxi Co. Ltd., Beijing, China) equipped with a TCD and a TDX-01 column for the separation of N₂O, N₂, and O₂. The N₂O conversion was designed as follows:

$$\mathrm{Conversion}\mathrm{of}{\mathrm{N}}_2\mathrm{O}(\%)=1-\frac{[{\mathrm{N}}_2\mathrm{O}]}{[{\mathrm{N}}_2\mathrm{O}]+[{\mathrm{N}}_2]}\times 100\%$$

(4)

where [N₂O] and [N₂] are the concentration of N₂O and N₂, respectively, as measured by the chromatograph. Additionally, the outlet gases of the reactor were further analyzed a five-way gas analyzer (SIEMENS VS5067-5D). This analyzer has an infrared optical sensor and is able to detect NO, CO, CO₂, O₂, and hydrocarbons (HCs). The detection result suggested that N₂O was directly decomposed into N₂ and O₂ without generation of NO_x over the investigated samples.

4. Conclusions

In the present work, a series of Mg-, Ce-doped Co₃O₄ composite spinel metal-oxide catalysts was prepared and evaluated for N₂O direct dissociation. It was found that the Mg, Co co-doped Co₃O₄ with an optimum Mg/Ce mole ratio of 0.5 (Mg_0.025Ce_0.05Co_0.925Co₂O₄) exhibited the highest N₂O conversion activity, which could achieve 100% N₂O conversion at T = 250 °C. Various characterizations using XRD, BET, H₂-TPR, and XPS were conducted to gain deeper insight into the Mg-, Ce-doping effect. Three main factors were found to contribute to the Mg_0.025Ce_0.05Co_0.925Co₂O₄ sample obtaining the highest deN₂O activity: (a) Mg and Ce co-doping could reduce the grain size of composite spinel metal oxide, resulting in a significant increase of the BET specific surface area from 32.5 g·m⁻² of pure Co₃O₄ to 111.2 g·m⁻²; (b) Mg and Ce co-doping could improve the redox ability of both Co³⁺ → Co²⁺ and Co²⁺ → Co²⁺; (c) Mg and Ce co-doping could improve the mobility as well as the concentration of surface atomic O.