Co-precipitation Synthesized MnOx-CeO2 Mixed Oxides for NO Oxidation and Enhanced Resistance to Low Concentration of SO2 by Metal Addition

Abstract

NO oxidation was conducted over MnO_x-CeO₂ catalysts, which were synthesized by the co-precipitation method. The calcination temperature and third metal doping were the main considerations. MnCe catalysts calcined at 350 °C and 450 °C attained the highest NO conversion efficiency, compared to 550 °C. XRD results suggested that the higher the calcination temperature, the higher the crystallization degree, which led to a negative effect on catalytic activity. Subsequently, Sn, Fe, Co, Cr, and Cu were separately doped into MnCe composites, but no improvement was observed for these trimetallic catalysts in NO conversion. Nevertheless, MnCeSn, MnCeFe, and MnCeCo still exhibited a desirable NO conversion efficiency, so they were tested under SO₂ addition together with MnCe catalyst. Among them, MnCeFe exhibited the highest NO conversion after whole poisoning testing. XPS results indicated that Fe could protect Mn and Ce metal oxides from being reduced during SO₂ poisoning process. Furthermore, in in-situ DRIFTS measurement, part of nitrate species maintained undestroyed on the MnCeFe catalyst surface after SO₂ poisoning. These characteristics reinforced that Fe dropping would achieve better performance under SO₂ atmosphere.

1. Introduction

BP’s Energy Outlook (2016) predicted that fossil fuels would still maintain the dominant sources of the world energy powering until 2035 [1]. Power plants, industrial boilers, and motor vehicle engine discharge NO_x via fossil fuels combustion inevitably [2]. After complex atmospheric chemical reactions, exhausted NO_x will be converted to the culprit of many typical environmental events, such as photochemical smog, ozone depletion, acid rain, and green-house effect [3,4,5]. Conventionally, the three-layer SCR (selective catalytic reduction) catalyst, assembled in coal-fired power plants, could satisfy the ultra-low emission standard in China (i.e., NO_x < 50 mg·Nm⁻³) [6]. However, the short slab of strict temperature window impedes the SCR application in industrial boilers, which produce steel, ceramic, glass, and cement, with relatively low temperature flue gas (i.e., <200 °C). Typically, 95% of NO_x in the exhaust gas is water-insoluble NO, but it can be easily be removed by wet scrubbing after oxidation to NO₂ [7,8]. Therefore, NO oxidation technology is considered an alternative for NO_x elimination, and it is also a critical step in SCR, LNT (lean NO_x trap), and NSR (NO_x storage-reduction) [9,10,11].

Transition metal oxides (TMO) are excellent candidates, given their rich various oxidation states, abundant oxygen defects, and earth rich features [12,13,14]. Among these catalysts, Mn-based, Ce-based, and Zr-based catalysts exhibit excellent performance at low temperature, and their SO₂ resistance was enhanced after the modification of other TMO [4,15,16,17]. In particular, the strong interaction between Mn and Ce created remarkable oxygen storage and redox ability [15,18], which contributed to extensive application in NO [19], as well as soot [20,21], mercury [22,23], VOCs [24,25] oxidation reactions. Lin et al. [26] synthesized Mn/CeO_x catalyst by the sol-gel method and reported ~90% NO conversion at 230 °C under the GHSV (Gas Hour Space Velocity) of 20,000 h⁻¹. Liu et al. [27] demonstrated that plasma assisted synthesis of Mn-Ce-O_x catalyst produced abundant Ce³⁺ species and active oxygen species, leading to better catalytic behavior. To further lower the efficient catalytic temperature, Shen et al. [28] employed carbon as a hard template to prepare a series of hollow MnO_x-CeO₂. Among them, the sample equipped with the largest surface area and pore volume values exhibited the best performance, whereby the maximum NO oxidation conversion (~82%) was obtained at 200 °C with 120,000 h⁻¹ GHSV. Recently, trimetallic catalysts, Mn-Co-Ce-O_x [1,15], Mn-Ce-Zr-O_x [19], Cu-Ce-Zr-O_x [29], attracted much attention in relation to mitigating SO₂ poisoning. However, the catalyst stability under SO₂ is still a hard nut to crack.

Considering industrial production, co-precipitation method has unique advantages in cost and process simplification compared with aforesaid sol-gel, plasma and template methods. In this paper, MnO_x-CeO₂ composites catalysts were synthesized by ammonium carbonate co-precipitation method. Sn, Fe, Co, Cr, Cu, the most commonly used TMO, were introduced to investigate their effect on activity and SO₂ resistance. Unfortunately, the oxidation atmosphere might have a positive effect on sulfuration [30], while no efficient method to avoid SO₂ poisoning exists. A shift in attention to low-sulfur flue gas is a better way to understand the SO₂ poisoning process in detail. In addition, the catalyst resistance under low concentration SO₂ has been rarely researched, which might have a broad application space in low-sulfur fuel combustion. Therefore, the investigation of low SO₂ resistance was carried out in this paper. XRD, XPS, BET and H₂-TPR were used to investigate the catalyst activity, and in-situ DRIFTS measurements were conducted to reveal the NO oxidation and SO₂ poisoning process.

2. Results and Discussion

2.1. XRD and BET Results

The XRD patterns of MnO_x-CeO₂ based catalysts, with different calcination temperatures and metal doping, are shown in Figure 1. All these catalysts consisted of diffraction peaks that corresponded to CeO₂ with a cubic fluorite structure. These diffraction peaks at 28.6°, 33.1°, 47.5°, 56.3°, 59.1°, 69.4°, 76.7°, 79.1°, and 88.4° are assigned to the crystalline planes of CeO₂ (111), (200), (220), (311), (222), (400), (331), (420), and (422) (JCPDS:34-0394) [31]. No diffraction peaks that correlated with MnO_x species were detected, indicating metals doping did not destroy the framework of CeO₂ [30,32,33]. The ionic radius of Mn³⁺ is 0.066, and Ce⁴⁺ is 0.1098 nm. Therefore, Mn³⁺ might incorporate into the fluorite lattice of CeO₂. Machida et al. [34] also found that Mn₂O₃ crystallization could be detected until Mn/(Mn+Ce) > 0.75, whereas it was 0.29 in this paper.

The diffraction peaks intensified with the increased calcination temperature, as shown in Figure 1a, which was consistent with the increasing crystallite size determined by Scherrer equation for (111) peak in Table 1. Clearly, increasing calcination temperature gave rise to a higher degree of crystallization, and resulted in lower catalytic activity. Figure 1b presents XRD patterns of MnO_x-CeO₂ with third metal doping. Obviously, no new diffraction peaks appeared after metal doping. The peaks of CeO₂ became broader and the intensity weakened significantly, especially for MnCeSn and MnCeCr. Also, their crystallite sizes reduced to 8.8 nm, and 9.9 nm, respectively (Table 1).

The textural properties of MnO_x-CeO₂ based catalysts are summarized in Table 2. Clearly, the surface area and pore volume of the trimetallic catalysts (MnCeSn, MnCeFe, and MnCeCr) were higher than that of the bimetallic catalyst (MnCe(450)). MnCeCo exhibited higher surface area, but lower pore volume compared with MnCe(450). However, the surface area and pore volume of MnCeCu were all smaller than MnCe(450). Among them, MnCeSn and MnCeCr possessed distinct increase in both surface area and pore volume, i.e., 106.6 m²·g⁻¹ and 0.19 mL·g⁻¹ for MnCeSn, 126.5 m²·g⁻¹ and 0.26 mL·g⁻¹ for MnCeCr. This was consistent with the obvious broader and weaker diffraction peaks of MnCeSn and MnCeCr in XRD patterns. Lower crystallization degree implied better pore structure, i.e., high surface area and pore volume. Generally, higher surface area and pore volume can provide more active adsorption sites for reactants, but these trimetallic catalysts didn’t exhibit fortunately higher activity. Especially, MnCeCr possessed the lowest NO oxidation activity in the activity test, even with the highest surface area and pore volume, indicating that the surface area is not the only factor for catalytic activity. This observation may be explained by their reducibility from TPR results next.

2.2. H₂-TPR Measurements

The H₂-TPR profiles of trimetallic catalysts are shown in Figure 2, and MnCe(450) was listed as a benchmark. Except for MnCeCu, there were two dominant reduction peaks for these catalysts, which were mainly assigned to MnO_x species three-step reduction: MnO₂→Mn₂O₃→Mn₃O₄→MnO [35,36]. Additionally, a weak tail peak, at a temperature higher than 400 °C, should be attributed to the reduction of the bulk Ce⁴⁺ to Ce³⁺ [37,38]. Obviously, the third metal was reduced at the similar reduction temperature range of MnO_x species. On the other hand, the third metal doping resulted in distorted H₂ reduction profiles, which led to the reduction profiles shifting to a higher temperature. These observations should be attributed to the metal interaction. Reducibility is a critical factor for NO oxidation. As the NO conversion curves of the activity test, the highest points were located near 250 °C. Therefore, the H₂ reduction profiles at temperature lower than or nearby to 250 °C are mainly correlated with the NO oxidation activity. Compared to MnCe(450), the first reduction peak at ~265 °C of the trimetallic catalysts was weaker, but the second peak at 330~400 °C intensified. As to MnCeCr, the two reduction peaks incorporated together, and the reduction temperature shifted to higher level entirely. These changes in reducibility after the third metal addition gave an explanation for their worse catalytic activity, even with better pore structure parameters (Table 2). The first H₂ reduction peak of MnCeSn, MnCeFe, and MnCeCo did not shift to a higher temperature, which was correlated with their relatively better activities. Due to the intrinsic, significantly low temperature reducibility of Cu, its doping resulted in the original MnO_x species reduction peaks moving to a lower temperature (<250 °C). The synergistic effect between CuO_x and MnO_x species contributed to more oxygen defects and structural distortion, then enhanced low temperature reducibility [39]. Nevertheless, the catalytic activity of MnCeCu still became worse compared with MnCe(450) in the activity test. As mentioned in Table 2, the surface area and pore volume of MnCeCu were the lowest among all the catalysts, which was unfavorable for reactants adsorption. On the other hand, a strong reducibility of Cu caused the reduction of Mn species [40]. NO oxidation proceeded through the MnO_x species valance state transformation between Mn⁴⁺ and Mn³⁺ [30,41]. Therefore, Mn ions reduction by Cu would inhibit NO oxidation reaction, and gave rise to the low catalytic activity of MnCeCu.

2.3. XPS Analysis

XPS characterization of MnCe(450) (abbreviated as MnCe) and MnCeFe was conducted to investigate the oxidation state of Mn, Ce, and the distribution of O species. The XP spectra of Mn 2p included two regions, i.e., Mn 2p_3/2 and Mn 2p_1/2. The region of Mn 2p_3/2 was deconvoluted into two peaks using Gaussian functions, as presented in Figure 3a. These two peaks were assigned to Mn³⁺ and Mn⁴⁺ [36,42] from low, to high binding energies, respectively. The binding energies and respective ratios estimated by integration are tabulated in Table 3. Notably, the binding energies shifted to higher value after Fe doping caused by electron transfer. The decrease in Mn⁴⁺ ratios demonstrates Mn ions reduction caused by Fe doping, which should be related to the lower activity in NO oxidation.

The Ce 3d spectra shown in Figure 3b exhibited complicated overlapping peaks and were deconvoluted into eight peaks after peak-fit processing. These peaks were assigned to two regions, i.e., Ce 3d_3/2 and Ce 3d_5/2, and the corresponding peaks were labelled as U and V series. The detailed binding energies and Ce⁴⁺ ratios are tabulated in Table 4. Two satellite peaks of U’ and V’ were assigned to Ce³⁺, and the other peaks were characteristic of Ce⁴⁺ [31,43]. The relative Ce⁴⁺ content in the catalyst was proportional to the crystallite size [44], and it was also supported by the XRD patterns. Ce³⁺ ions represent reduced, non-stoichiometric cerium species. The co-existence of Ce⁴⁺ and Ce³⁺, as well as Mn⁴⁺ and Mn³⁺, provided electron exchange for the redox reactions, and then contributed to the catalytic activity in NO oxidation. Notably, the ratio of Ce⁴⁺/Ce increased from 56.2% (MnCe) to 65.9% (MnCeFe) after Fe doping. Overall, Ce³⁺ was oxidized into higher oxidation state, and Mn⁴⁺ was reduced into lower oxidation state after Fe doping into MnCe mixed oxides.

The O 1s XP spectra are presented in Figure 3c that are classified into two main bands. After the deconvolution process, two or three characteristic peaks were observed, with the corresponding binding energies tabulated in Table 5. From low to high binding energies, these peaks were assigned to O_α, O_β, and O_γ, respectively. O_α represents lattice oxygen, while the surface adsorbed oxygen, hydroxyl groups and carbonates are associated with O_β, and O_γ [31], respectively. The distribution ratios of oxygen species are listed in Table 5. Clearly, the ratio of O_β and O_γ increased from 26.3% and 8.9% (MnCe) to 33.6% and 14.8% (MnCeFe), demonstrating more surface oxygen species was created after Fe addition.

2.4. Catalytic Activity

2.4.1. MnO_x-CeO₂ with Different Calcination Temperature

Carbonate composites of Ce₂(CO₃)₃ and Mn(CO₃)₂ were formed after the co-precipitation process via Equations (1) and (2). Composite oxides of MnO_x and CeO₂ were then produced via carbonate decomposition after the calcination process. To investigate the effect of calcination temperature on NO oxidation, three catalysts calcined at 350 °C, 450 °C, and 550 °C were prepared, and the corresponding catalytic activities are shown in Figure 4. The dot line is the thermodynamic limitation curve for NO oxidation. MnCe(350) and MnCe(450) exhibited similar NO conversion within the whole temperature range, whereas a distinct decrease in NO conversion was observed for MnCe(550). MnCe(350) and MnCe(450) achieved the highest NO conversion efficiency of ~88% at 220~250 °C. But, only 83% was obtained as the highest value for MnCe(550) at 280 °C. Therefore, the calcination temperature was a significant synthesis factor for NO oxidation, and 350~450 °C was a comparatively suitable range for MnCe catalyst calcination. Table 6 illustrated the NO oxidation activities of metal-oxides in literatures. It can be seen that the MnCe catalyst in our paper has a relatively high NO conversion efficiency at low temperature:

2Ce(NO₃)₃ + 3(NH₄)₂CO₃ → Ce₂(CO₃)₃ + 6NH₄NO₃.(1)

Mn(CH₃COO)₂ + (NH₄)₂CO₃ → MnCO₃ + 2CH₃COONH₄.(2)

2.4.2. MnO_x-CeO₂ with Metal Doping

MnCe(450) was used as the benchmark catalyst, while Sn, Fe, Co, Cr, and Cu was added separately into MnO_x-CeO₂ composite oxides as the doping metal, and their catalytic activities are shown in Figure 5. Unfortunately, no improvement could be observed for all these doping catalysts, whereas different degrees of decrease are presented. Similar NO conversion curves were presented for MnCeSn, MnCeFe, and MnCeCo. They obtained the highest NO conversion efficiency of ~85%, but it was first observed until 250 °C. For MnCeCu and MnCeCr, the NO conversion curves further shifted to higher temperature. MnCeCr exhibited the worst catalytic activity, i.e., the highest value was 70% at 310 °C.

2.5. Catalyst Resistance to SO₂

2.5.1. Variation of NO Oxidation with SO₂ Addition

To investigate the effect of SO₂ on catalytic activity in NO oxidation, low concentration of SO₂, i.e., 20 ppm, was introduced into a catalytic reactor to observe the slow poisoning process. SO₂ poisoning was investigated on MnCe, MnCeFe, MnCeSn, and MnCeCo catalyst. The NO conversion curves at 250 °C along with time on stream are presented in Figure 6. The 13 h testing can be divided into four sections, i.e., temperature ramp to 250 °C within 2 h, stabilization within 100 min, SO₂ addition with staged concentrations (20~40~20 ppm), and non-SO₂. The concentrations of NO and NO₂ in the entire process are also recorded for comparison. A similar tendency of NO and NO₂ concentration curves can be observed in the temperature ramp section in Figure 6a–d. Initially, NO concentration increased to a specific value. Next, NO concentration drastically decreased after a previous increase, which is likely related to the promoted adsorption at a certain temperature. The sharp decline was found between two peaks for NO concentration curves. Meanwhile, NO₂ appeared right on the heels of the NO concentration valley bottom. After saturation adsorption, just at the valley bottom position, further increased temperature caused desorption of NO and NO₂. When the temperature reached a starting value for NO oxidation, NO concentration decreased again, whilst NO₂ concentration further increased until stabilization.

The detailed NO conversion efficiency values, at four critical points, are summarized in Table 7. After 100 min stabilization, NO conversion efficiencies of these catalysts all reached their highest values. When SO₂ was injected into the reactor, NO conversion of MnCe maintained the highest value for 166 min, and then decreased to 77% at the end of the first 360 min. However, unlike MnCe, the NO conversion efficiency of MnCeFe, MnCeSn, and MnCeCo decreased immediately once 20 ppm SO₂ addition, reaching to 76%, 68%, and 70% after 360 min test, respectively. Compared with the decrement from 86% to 76% of MnCeFe, the descending rate for MnCe seems to be relatively steeper from 90% to 77%. Interestingly, NO concentration always increased later than the decrease of NO₂ concentration. This phenomenon was ascribed to the NO₂ desorption expelled by sulfate formation on the catalyst surface, resulting in a relatively delayed NO₂ concentration. Almost no reduction in NO₂ concentration was observed for MnCeFe during 360-min SO₂ poisoning testing (20 ppm), indicating excellent NO₂ storage capacity. Subsequently, SO₂ concentration was increased to 40 ppm. Clearly, the decline in NO conversion sped up, especially for MnCe, i.e., which reduced 23%, from 77% to 54% in 60 min. According to the data in Table 7, the total decrement within this 60-min poisoning process of MnCeFe, MnCeSn, and MnCeCo were 7%, 11%, and 12%, respectively. This indicates trimetallic catalysts improved the resistance to SO₂ with relatively higher concentration. When the SO₂ concentration was switched back to 20 ppm, the decline rate was alleviated except for MnCeFe. It is worth mentioning that the decrement of NO conversion efficiency for MnCe was unexpectedly the highest, although it was the latest one to start decreasing in the first 360-min poisoning. NO conversion efficiency remained stable when SO₂ was dislodged from simulated flue gas for all the four catalysts. Overall, MnCeFe exhibited the highest resistance to SO₂, which possessed 62% NO conversion efficiency after whole poisoning testing.

2.5.2. Surface Properties after SO₂ Poisoning Process

The oxidation states of Mn, Ce, Fe, and the distribution of O species before, and after, the SO₂ poisoning process were investigated by XPS characterization. The XP spectra of Mn 2p, Ce 3d, and O 1s are presented in Figure 3, with the detailed binding energies and ratios tabulated in Table 3, Table 4 and Table 5. Clearly, the most distinct changes occurred in the O 1s spectra, i.e., the lattice oxygen (O_α) decreased and transformed into surface adsorbed oxygen species (O_β and O_γ) after SO₂ poisoning process. This observation should be mainly attributed to the accumulation of sulfate species on the catalyst surface [30]. Additionally, changes in Mn 2p and Ce 3d were different for MnCe and MnCeFe. The ratio of Mn⁴⁺ of MnCe decreased from 51.1% to 41.5% after the SO₂ poisoning process; whereas, it increased a little from 36.5% to 39.9% for MnCeFe. In terms of Ce 3d, these two catalysts all exhibited increase in Ce⁴⁺ ratio after SO₂ addition. Generally, the formation of sulfate species on catalyst surface would reduce the metal oxides oxidation state. For MnCe catalyst, reduction of metal oxidation state was mainly occurred in Mn ions. However, no reduction was found in both Mn ions and Ce ions for MnCeFe catalyst. It can be speculated that Fe ions took this responsibility instead of Mn. Therefore, Fe 2p XP spectra were carried out, shown in Figure 7, and it was deconvoluted into four peaks using Gaussian functions. These peaks were assigned to Fe³⁺, Fe²⁺ and satellite peaks from low to high binding energy. The corresponding binding energies and ratios of Fe ions are tabulated in Table 8. Obviously, Fe³⁺ decreased from 41.2% to 21.9% accompanied by the increase of Fe²⁺. Herein, Fe played a protective role in the reduction of Mn and Ce when SO₂ was added, and as a result, the SO₂ resistance was enhanced.

2.5.3. In-Situ DRIFTS Measurement

In-situ DRIFTS measurements were conducted, in order to investigate the surface species formation during NO oxidation and the SO₂ poisoning process. The infrared absorption spectra of NO (200 ppm)-O₂ (10%) co-adsorption and NO-O₂-SO₂ co-adsorption at 250 °C were collected successively, as listed in Figure 8 and Figure 9. Firstly, diverse absorption bands emerged after NO and O₂ injection, which were assigned to nitrate and nitrite species. For MnCe catalyst, shown in Figure 8a, the dominated bands at 1311 cm⁻¹, 1382 cm⁻¹, and 1473 cm⁻¹ were associated to bridged nitrate [50,51], adsorbed NO₃⁻ [52], and cis-N₂O₂²⁻ [32], respectively. Several weak bands (903 cm⁻¹, 832 cm⁻¹, and 813 cm⁻¹), lower than 1043 cm⁻¹, should be assigned to nitrate species on manganese and cerium oxides [32]. All the appeared peaks increased quickly at first 30 min, and then maintained at a certain level, indicating saturation adsorption. These nitrate and nitrite species on the catalyst surface were considered as the critical intermediates for NO oxidation. After 60-min co-adsorption of NO-O₂, SO₂ was added, and the infrared absorption spectra are presented in Figure 8b. In the first 40 min, only 20 ppm SO₂ was added. The band at 1114 cm⁻¹ increased first, which corresponded to the bulk-like bidentate sulfate species [53,54,55,56], and developed into the dominated band finally. Subsequently, the bands at 974 cm⁻¹, 1029 cm⁻¹, and 1277 cm⁻¹ began to be observed. However, they were not intensified strikingly along with time. These accessory bands were assigned to either surface or bulk-like sulfate species [57]. As SO₂ concentration further increased, the new bands became the most prominent bands on the whole spectrum, demonstrating that S species conquered N species and took up the active sites. Fortunately, the bands at 1382 cm⁻¹ and 1473 cm⁻¹ were still reserved to launch NO oxidation.

The in-situ DRIFTS measurements under the same conditions were conducted over MnCeFe catalyst, shown in Figure 9. NO-O₂ co-adsorption resulted in N species accumulation on catalyst surface, corresponding to these bands at 1270 cm⁻¹, 1310 cm⁻¹, 1360 cm⁻¹, and 1466 cm⁻¹ observed in Figure 9a. Interestingly, compared with MnCe in Figure 8a, the new band at 1270 cm⁻¹ (the bidentate nitrate), did not disappear under SO₂ impact in Figure 9b, and even intensified along with time. Also, no shift on the band position was detected. Meanwhile, the other bands at 1360 cm⁻¹ and 1466 cm⁻¹ shifted to a little higher wave-number, but maintained its intensity. It can be speculated that the positive SO₂ resistance performance of MnCeFe could be related to these undestroyed N species on catalyst surface. Not surprisingly, the bands linked to S species were also intensified. Unlike MnCe, the band at 1118 cm⁻¹ was not the dominated one anymore. Other bands at 1029 cm⁻¹ and 1230 cm⁻¹ also occupied almost equal status.

3. Experiments and Methods

3.1. Catalyst Preparation

The MnO_x-CeO₂ catalysts were synthesized by the co-precipitation method. A total of 0.01 mol Ce(NO₃)₃·6H₂O (AR, 99.0%, Sinopharm, Shanghai, China) and 0.005 mol Mn(CH₃COO)₂·4H₂O (AR, 99.9%, Aladdin) were dissolved together in 75 mL distilled water. Subsequently, an excess proportion of (NH₄)₂CO₃ (AR, ≥40.0%, Sinopharm, Shanghai, China) solution (120 mL) was instilled into the mixed Ce-Mn precursor solution with a flow rate of 10 mL·min⁻¹ under vigorous stirring. After continuously stirring for 2 h, the mixture was aged for another 3 h at room temperature. The precipitate was obtained by centrifugation and washed with deionized water several times. Finally, the product was dried at 110 °C overnight and then calcined in a tube furnace at a target temperature for 3 h with a ramp rate of 1 °C·min⁻¹. Three MnO_x-CeO₂ catalysts were calcined at 350 °C, 450 °C, and 550 °C, respectively, to investigate the calcined temperature effect on catalytic activity. The catalysts were denoted as MnCe(350), MnCe(450), and MnCe(550), respectively. As for the trimetallic catalysts, 0.001 mol SnCl₄·7H₂O (AR, 99.0%, Aladdin), Fe(NO₃)₃·9H₂O (AR, 98.5%, Sinopharm, Shanghai, China), Co(NO₃)₂·6H₂O (AR, 98.5%, Sinopharm, Shanghai, China), Cr(NO₃)₃·9H₂O (AR, 99.0%, Sinopharm, Shanghai, China), and Cu(NO₃)₂·3H₂O (AR, 99.0%, Sinopharm, Shanghai, China) were separately dissolved together with Mn-Ce precursors above, while the Mn(CH₃COO)₂·4H₂O was reduced to 0.004 mol. The rest of the preparation procedure was the same with the former MnCe, and the calcination temperature was fixed at 450 °C. The catalysts were denoted as MnCeSn, MnCeFe, MnCeCo, MnCeCr, and MnCeCu, respectively.

3.2. Activity and Stability Tests

The activity and stability tests of MnO_x-CeO₂ catalysts were conducted in a fixed-bed stainless-steel-tube reactor with an 8 mm internal diameter. A K-type thermocouple was placed in direct contact with the sample to detect the real catalytic temperature. The reaction gas was supplied by cylinder gas (N₂- 99.999%, O₂- 99.999%, NO-5%/balance N₂, SO₂-5%/balance N₂), with the total flow rate of 240 mL·min⁻¹. The initial NO concentration was 200 ppm, and 10% O₂ was injected separately to the reactor as the oxidant. All the samples were sieved to 40~60 mesh and then blended with a moderate amount of quartz sand. A 0.2 g sample was placed in the reactor center with the gas hour space velocity (GHSV) of ~15000 h⁻¹. SO₂ was introduced into the reactor at 250 °C after 100-min stabilization to investigate catalyst resistance to SO₂. NO conversion efficiency was calculated according to the outlet NO and NO₂ concentrations, which were constantly detected by a Fourier transform infrared gas analyzer (Gasmet FTIR DX4000, Vantaa, Finland), as shown in the Equation (1):

[Conv.]_NO = [NO₂]_out/([NO]_out + [NO₂]_out) ×100 % (3)

where [Conv.]_NO is the NO conversion efficiency, [NO₂]_out in ppm is the outlet NO₂ concentration of the reactor, and [NO]_out in ppm is the outlet NO concentration of the reactor.

3.3. Catalyst Characterization

The XRD (X-ray diffraction) patterns were detected by using a Rigaku D/MAX-2500 diffractometer (Rigaku Co., Tokyo, Japan). N₂ adsorption-desorption isotherms were measured in a Micromeritics ASAP 2010 analyzer (Micromeritics Instrument Corp, Norcross, GA, USA) at 77 K. The XPS (X-ray photoelectron spectroscopy) was measured through a photoelectron spectrometer (Thermo Scientific Escalab 250Xi, Thermo Fisher Scientific, Waltham, MA, USA) with a standard Al Kα source (1486.6 eV). H₂-TPR measurements were conducted at an automatic temperature programmed chemisorption analyzer (Micromeritics AutoChem II 2920, Micromeritics Instrument Co., Norcross, GA, USA). ~50 mg of catalyst was loaded for TPR tests under the atmosphere 30 mL·min⁻¹ 10% H₂/Ar after purged by He (>99.99%) at 120 °C for 1 h. The typical TPR process was started from 100 °C to 800 °C by a ramp of 10 °C·min⁻¹. A Nicolet iS50 FTIR spectrometer (Thermo Nicolet, Madison, WI, USA) equipped with an MCT/A detector was used for in-situ DRIFTS (Diffused Reflectance Infrared Fourier Transform Spectroscopy) experiments. Prior to each test, all samples were pretreated by 80 mL·min⁻¹ of N₂ at 120 °C for 1 h. Then, the in-situ spectra were recorded from 600 to 4000 cm⁻¹ at 250 °C under specific flue gas. The catalysts after SO₂ poisoning were denoted as MnCe_S and MnCeFe_S, respectively.

4. Conclusions

MnO_x-CeO₂ catalysts were synthesized (molar ratio of Mn/Ce = 0.5) by co-precipitation method to investigate their catalytic activity of NO oxidation. Metal oxides were generated through the thermal decomposition of carbonate composites after the co-precipitation process. Then, the samples were calcined at different temperatures, i.e., 350°C, 450 °C and 550 °C, to evaluate the calcination temperature effect. Subsequently, the third metal (M), i.e., Sn, Fe, Co, Cr, and Cu, was doped separately into MnCe composites to attain trimetallic catalysts with 0.1 M/Ce and 0.4 Mn/Ce molar ratios for NO conversion and SO₂ resistance test. Several conclusions are listed below combined with characterization results.

• (1). Higher calcination temperatures brought higher crystallization degree. MnCe catalysts calcined at 350 °C and 450 °C exhibited higher NO conversion than that calcined at 550 °C.

• (2). The third metal doping (Sn, Fe, Co and Cr) could reduce crystallization degree, and then improved the surface area or pore volume, but inhibited low temperature reducibility in H₂-TPR, except for MnCeCu.

• (3). For SO₂ poisoning, Fe addition into MnCe catalyst could protect Mn and Ce metal oxides from being reduced. In DRIFTS measurement, the decomposition of surface nitrates at SO₂ presence gave evidence to the decreasing NO conversion efficiency. Additionally, the undestroyed parts of nitrate species on the MnCeFe catalyst surface, after SO₂ poisoning, contributed to its better SO₂ tolerance.

Author Contributions

Conceptualization, Z.W. and F.L.; methodology, Z.W.; software, J.S. and Y.L.; validation, Z.W., F.L. and G.C.; investigation, J.S. and H.T.; resources, Y.L.; data curation, J.S. and P.L.; writing—original draft preparation, J.S. and F.L.; writing—review and editing, Z.W., J.S., F.L. and G.C.; funding acquisition, Z.W. All authors read, revised and approved the manuscript.

Funding

This research was funded by the National Key Research and Development Program of China, grant number 2018YFB0605200, and Zhejiang Provincial Natural Science Foundation, grant number LR16E060001.

Acknowledgments

This work was supported by the National Key Research and Development Program of China, grant number 2018YFB0605200, Zhejiang Provincial Natural Science Foundation, grant number LR16E060001, and National Natural Science Foundation of China, grant number 51621005.

Conflicts of Interest

The authors declare no conflict of interest.

Figures and Tables

Figure 1. XRD patterns of MnOx-CeO2 based catalysts, (a) MnOx-CeO2 with different calcination temperature, (b) MnOx-CeO2 with metal doping.

Figure 2. H2-TPR profiles of MnCe(450), MnCeSn, MnCeFe, MnCeCo, MnCeCr, and MnCeCu.

Figure 3. XP spectra of MnCe, MnCe_S, MnCeFe, and MnCeFe_S, (a) Mn 2p, (b) Ce 3d, (c) O 1s.

Figure 4. NO conversion over MnOx-CeO2 with different calcination temperature.

Figure 5. NO conversion over MnOx-CeO2 with metal doping.

Figure 6. Effect of SO2 on NO conversion efficiency at 250 °C: (a) MnCe, (b) MnCeFe, (c) MnCeSn, and (d) MnCeCo.

Figure 6. Effect of SO2 on NO conversion efficiency at 250 °C: (a) MnCe, (b) MnCeFe, (c) MnCeSn, and (d) MnCeCo.

Figure 7. Fe 2p XP spectra of MnCeFe and MnCeFe_S.

Figure 8. In-situ DRIFTS spectra of MnCe during NO–O2 co-adsorption ((a) 200 ppm NO–10% O2, N2 balance and (b) 20~40~80 ppm SO2 addition).

Figure 9. In-situ DRIFTS spectra of MnCeFe during NO–O2 co-adsorption ((a) 200 ppm NO–10% O2, N2 balance and (b) 20~40~80 ppm SO2 addition).

Table 1

Crystallization properties of MnO_x-CeO₂ based catalysts.

Catalyst	2θ(111)/°	FWHM	Crystallite Size a/nm
MnCe(350)	28.6	0.902	9.9
MnCe(450)	28.4	0.729	11.6
MnCe(550)	28.5	0.737	12.1
MnCeSn	28.6	1.017	8.8
MnCeFe	28.6	0.742	12.1
MnCeCo	28.6	0.772	11.6
MnCeCr	28.7	0.901	9.9
MnCeCu	28.5	0.792	11.3

^a Crystalline size determined from Scherrer equation for (111) peak.

Table 2

Textural properties of MnO_x-CeO₂ based catalysts.

Catalyst	BET Surface Area/m2·g−1	Pore Volume a/mL·g−1	Avg. Pore Diameter b/nm
MnCe(450)	73.3	0.10	9.8
MnCeSn	106.6	0.19	14.1
MnCeFe	111.7	0.11	5.5
MnCeCo	95.9	0.08	5.0
MnCeCr	126.5	0.26	14.3
MnCeCu	63.8	0.07	7.7

^a Single point adsorption total pore volume of pores less than 40.3 nm diameter at P/P₀ = 0.95. ^b BJH desorption average pore diameter.

Table 3

2p XPS parameters of MnCe, MnCe_S, MnCeFe, and MnCeFe_S.

Catalyst	Mn3+		Mn4+
	B.E. (eV)	Mn3+/Mnn+ (%)	B.E. (eV)	Mn4+/Mnn+ (%)
MnCe	640.9	48.9	642.5	51.1
MnCe_S	641.2	58.5	644.1	41.5
MnCeFe	641.1	63.5	643.4	36.5
MnCeFe_S	641.4	60.1	643.9	39.9

Table 4

Ce 3d XPS parameters of MnCe, MnCe_S, MnCeFe and MnCeFe_S.

Catalyst	Ce4+						Ce3+		Ce4+/Ce
	V	V’’	V’’’	U	U’’	U’’’	V’	U’	%
MnCe	881.7	888.2	897.7	900.2	906.9	916.2	883.0	900.4	56.2
MnCe_S	882.0	888.4	897.9	900.4	907.2	916.6	883.1	900.5	57.2
MnCeFe	881.9	888.3	897.9	900.4	906.9	916.3	883.2	902.0	65.9
MnCeFe_S	882.2	888.6	898.2	900.7	907.0	916.6	883.4	902.3	72.1

Table 5

O 1s XPS parameters of MnCe, MnCe_S, MnCeFe, and MnCeFe_S.

Catalyst	Oα		Oβ		Oγ
	B.E. (eV)	Oα/O (%)	B.E. (eV)	Oβ/O (%)	B.E. (eV)	Oγ/O (%)
MnCe	528.8	64.8	530.9	26.3	532.8	8.9
MnCe_S	529.2	32.6	530.5	3.1	532.1	64.3
MnCeFe	529.0	51.6	531.2	33.6	533.1	14.8
MnCeFe_S	529.3	39.4	531.9	60.6	--	--

Table 6

Summary of the metal oxides catalytic activities in literatures.

Catalyst	Preparation Method	Reaction Condition	NO Conversion	Ref
MnCe	Co-precipitation	200 ppm NO, 10% O2, 15,000 h−1, 220 °C	88%	This paper
MnCeOx	Citric acid method	500 ppm NO, 3% O2, 50,902 h−1,150 °C	46%	[45]
MnCeOx	Sol-gel method	400 ppm NO, 5% O2, 360,000 h−1, 300 °C	65%	[27]
MnCeOx	Co-precipitation	200 ppm NO, 8% O2, 5% H2O, 8% CO2, 25,000 h−1, 350 °C	69%	[19]
MnCeOx	CS template method	250 ppm NO, 5% O2, 120,000 h−1, 220 °C	82%	[28]
CrCeOx	Hydro-thermal method	400 ppm NO, 8% O2, 35,400 h−1, 300 °C	66%	[46]
CeCoOx	Sol-gel method	300 ppm NO, 10% O2, 20,000 h−1, 230 °C	93%	[26]
SmMn2O5	Co-precipitation	400 ppm NO, 10% O2, 100,000 h−1, 330 °C	52%	[47]
Cu/Ce0.8Zr0.2O2	Deposition-precipitation	500 ppm NO, 5% O2, 80,000 h−1, 320 °C	73%	[29]
MnCeCoOx	Co-precipitation	500 ppm NO, 3% O2, 35,000 h−1, 150 °C	80%	[48]
CoZrCeOx	Citrate complexation	3900 ppm NO, 8% O2, 30,000 h−1, 300 °C	80%	[49]

Table 7

NO conversion efficiency values at four critical points.

Catalyst	Highest Value	360-min (20 ppm SO2)	60-min (40 ppm SO2)	60-min (20 ppm SO2)
MnCe	90%	77%	54%	42%
MnCeFe	86%	76%	69%	62%
MnCeSn	86%	68%	57%	51%
MnCeCo	87%	70%	58%	53%

Table 8

Fe 2p XPS parameters of MnCeFe and MnCeFe_S.

Catalyst	Fe3+		Fe2+
	B.E. (eV)	Fe3+/Fe (%)	B.E. (eV)	Fe2+/Fe (%)
MnCeFe	709.9	41.2	712.0	58.8
MnCeFe_S	710.1	21.9	711.1	78.1