Stability Mechanism of Low Temperature C2H4-SCR with Activated-Carbon-Supported MnO x -Based Catalyst.

Manganese-based catalysts have shown great potential for use as a hydrocarbon reductant for NO x reduction (HC-SCR) at low temperatures if their catalytic stability could be further maintained. The effect of CeO2 as a promoter and catalyst stability agent for activated carbon supported MnO x was investigated during low temperature deNO x based on a C2H4 reductant. The modern characterization technology could provide a clear understanding of the activity observed during the deNO x tests. When reaction temperatures were greater than 180 °C and with ceria concentrations more than 5%, the overall NO conversion became stable near 70% during long duration testing. In situ DRIFTS shows that C2H4 is adsorbed on the Mn3Ce3/NAC catalysts to generate hydrocarbon activated intermediates, R-COOH, and the reaction mechanism followed the E-R mechanism. The stability and the analytical data pointed to the formation of stable oxygen vacancies within Ce3+/Ce4+ redox couplets that prevented the reduction of MnO2 to crystalline Mn2O3 and promoted the chemisorption of oxygen on the surface of MnO x -CeO x structures. Based on the data, a synergetic mechanism model of the deNO x activity is proposed for the MnO x -CeO x catalysts.

Introduction

Large amounts of NO emissions into the atmosphere lead to photochemical smog and chemical and eutrophication.[1−3] Hence, NO emission standards are becoming more rigorous, and NO reduction technologies, such as NO direct decomposition, selective noncatalytic reduction (SNCR), and selective catalytic reduction (SCR), have become more intensely investigated;[4−6] of these, SCR is the most diffusely used technology to control NO emission. The most established, commercial strategy for abating NO emissions during SCR is to use NH3 as the reductant,[7−10] and commercial catalysts are based mainly on V2O5–WO3/TiO2.[11−13] Ammonia is expensive; requires special handling, transport, storage, and application systems; and requires sophisticated metering to minimize NH3 slip.[14,15] NH3 slip causes plugging of ducts downstream of air-preheaters, forms secondary, fine particulate that is emitted into the atmosphere, and is regarded as an air pollutant potentially more objectionable than NO.[16,17] In addition, spent vanadium-based catalyst is classified as a detrimental solid waste, and its post processing is also costly.[18,19]

In contrast with NH3, hydrocarbons (HC) like C2H4 and C3H8 have low cost, high reducibility, user-friendliness, widespread distribution, and no NH3 slip, and most importantly, they are byproducts in the refinery station,[20] so they can be directly used to reduce the NO in the refinery station instead of long distance transportation and storage. Furthermore, hydrocarbons (HC) do not suffers from fouling by salt deposition, and some refinery stations have already used hydrocarbons (HC) to reduce NO. The overall Hydrocarbon-SCR reaction equation is as follows:

In addition, vanadium-based NH3–SCR operates in a narrow temperature window of 350–420 °C.[13] Hence, they are always used ahead of dust catcher and desulfurization systems to fulfill the desired temperature windows without extra heating. However, the presence of SO2 and dust in these process locations reduces catalyst performance and durability.[21,22] Nevertheless, if the running temperature could be reduced to, for instance, less than 250 °C, then de-NO facilities could be installed downstream of heat recovery steam generators (HRSG). Importantly, small-scale industrial boilers, such as steel furnaces, glass kilns, and coke ovens, operate with flue gas temperatures about 200–300 °C[23−25] and temperatures of desulfurization units below 200 °C. Under these conditions, vanadium-based catalysts would not be effective.[26,27] Thereby, developing effective and stable catalysts using low-temperature (<200 °C) HC-SCR has become a scientific and development quest with important practical implications.[24]

Among various catalyst active components investigated for HC-SCR of NO, noble metals have shown promising results;[28] for example, an Ag–Mg–Al composition (Ag7MgAl) at 350 °C exhibited 98% NO conversion. In general, however, noble metals have high costs, limiting their application in large-scale commercial operations; they are also not applicable for low-temperature HC-SCR reactions because their reactivities require temperatures near or higher than 350 °C. With concerns of low-temperature operating conditions, transition metal oxides served as potential candidates for HC-SCR catalysis because of their reactivity, low cost, and environmental acceptance.[29,30] For example, NO conversion was reported while using Cu/ZSM-5 with C3H6 as a reductant;[31] MnO is known for its initial, high reactivity at low temperatures for HC-SCR reactions.[32] Supports, such as Al2O3,[33] TiO2,[34] SiO2,[35] and activated carbon (AC),[36] have been used with MnO-based catalysts. Both SiO2 and TiO2 have demonstrated thermostability and acceptable mechanical capacities, while AC presents high surface area and porosity that imply the possibility of readily dispersing the active metal oxide and enabling reactant access to the active sites;[37,38] however, most of these studies have used NH3 as the reductant.[39]

MnO/AC[40] was used with C2H4 as a reductant and exhibited over 90% NO conversion even at temperatures as low as 130 °C; it deactivated rapidly such that only 20% conversion was evident after only 2 h of testing. This deactivation was attributed to a combination of the following: first, the valence state of the Mn decreased from Mn4+ to Mn3+ and Mn2+, but Mn4+ was found to be more active than the other two valences; second, carbon black was deposited on the active carbon surface during SCR reaction testing, blocking channels, and active catalytic sites. Cerium oxides (CeO2) have a specific fluorite structure, in which with each Ce4+ is surrounded by eight equivalent, nearest O2– ions which formed the corners of a cube structure and coordinated to four Ce4+ ions.[41] The ion configuration contributes to a high oxygen storage capacity, a vast number of oxygen vacancies, and strong redox properties between Ce3+ and Ce4+, which may help maintain the stability of MnO2. Therefore, ceria as an additive was considered a strong candidate to help promote and maintain the catalytic performance of MnO-based catalysts during HC-SCR reaction testing at low temperatures.[42−44] Characterization data shows that there were a number of oxygen defects produced by Mn3+ + Ti4+ ↔ Mn4+ + Ti3+ and Ce4+ + Mn3+ ↔ Ce3+ + Mn4+, which was a beneficial improvement of deNO, these two reactions generated a charge imbalance, oxygen vacancies, and unsaturated chemical bonds on the catalyst surface.[45,46]

This study will investigate the catalytic performance of ceria modified Mn/AC catalysts, and further explore the promotion mechanism of CeO2, which could help develop stable and highly active HC-SCR catalysts. The effect of CeO2 as a promoter and a catalyst stability agent for activated carbon supported MnO will be investigated during low temperature deNO based on C2H4 reductant. The modern characterization technology could provide a clear understanding of the activity observed during the deNO tests.

Results and Discussion

Dispersion of Active Component

SEM images and EDS elemental mappings of fresh Mn3Ce3/NAC catalyst are presented in Figure . The catalyst morphology was similar to that in previous studies,[40] the active components are evenly loaded on the surface of the activated carbon with particle structure. Elemental distributions indicated well-dispersed manganese and cerium on the NAC surface and without agglomeration,[47] which helps to improve the catalytic efficiency of the catalyst.

Figure 1

SEM and EDS images of Mn3Ce3/NAC: (a) SEM image Mn3Ce3/NAC SEM, 5000×; (b) EDS mapping of O; (c) EDS mapping of Mn; (d) EDS mapping of Ce.

Catalytic Performance

Figure a shows the HC-SCR catalytic performance of Mn3Ce3/NAC at 150, 180, and 200 °C. Similar to MnO/NAC without CeO,[20] excellent NO conversion was observed at 150 and 180 °C during the first 20 min of testing, but large decreases in NO conversion then occurred, such that only about 20% NO conversion remained after 2 h of reaction testing. In contrast, at 200 °C after an initial sharp drop in conversion, the activity was constant at 65% over the entire 2 h period of testing. This may be because at the initial stage of the reaction, most of Mn and Ce in the fresh catalyst are in the forward direction in the catalytic reduction reaction, but the reaction is not balanced at this time. After the chemical reaction is in equilibrium, the NO conversion rate tends to be stable, and the valence state changes of Mn and Ce in the catalyst also tend to be stable. N2O less than 2 ppm was measured by GC to, so it was not discussed in this study. The N2 selectivity of Mn3Ce3/NAC is shown in Figure b; as can be seen, it could remain at a high level for above 90%, especially for reaction at 200 °C, where the N2 selectivity reached 100% during overall reaction time.

Figure 2

(a) NO conversion. (b) N2 selectivity of Mn3Ce3/NAC using C2H4 as a reductant. Reaction temperature varied from 150 to 200 °C; reaction conditions were 500 ppm of NO, 250 ppm of C2H4, and 3% O2 with a flow rate of 1500 mL/min.

Effects of the Ce loading on NO conversion and N2 selectivity at 180 °C were investigated (Figure ). The NO conversion times for Mn3Ce3/NAC and Mn3Ce5/NAC were very similar to the results from Mn3Ce3/NAC as shown in Figure a when the reaction temperatures were either 150 or 180 °C. However, when the Ce loading was increased to 7% and the reaction temperature was 180 °C, the NO conversion rate remained stable over the entire period of the testing. By comparing the data in Figure a and Figure a, it was concluded that both reaction temperatures and CeO concentrations can affect NO conversion rates and stabilities during HC-SCR.

Figure 3

(a) NO conversion and (b) N2 selectivity of Mn3Ce/NAC (x = 3, 5, 7) used C2H4 as a reductant. Reaction temperature was 180 °C, and the reaction conditions were 500 ppm of NO, 250 ppm of C2H4, and 3% O2 with a flow rate of 1500 mL/min.

The presence of CeO has been shown to be capable of mitigating the deactivation of Mn/NAC catalysts during low temperature HC-SCR reaction testing.[40] Hence, physical and chemical differences in MnCe/NAC catalysts were investigated before and after reaction testing to shed light on changes which may affect NO conversion rates.

Promoting Mechanism of Ce Loading

Effect of Ce Addition on the Surface Area

As displayed in Table , the BET surface areas were decreased by about 15% upon impregnating CeO onto a MnO/NAC catalyst, but its surface area was not significantly different after reaction testing; these results suggest that MnO and CeO had minimal influence on the textural properties and the surface structure of the catalyst.[48] Although the larger surface area is known to be beneficial to the chemical reactivity of HC-SCR reactions,[49] it was not the key to affecting the catalytic activity of Mn3Ce3/NAC, because the decrease of catalytic performance was more drastic than the change of surface area.

Table 1

Specific Surface Area

samples	BET surface area (m2/g)
Fresh AC	439.07
Mn3 /NAC Before Reaction	668.54
Mn3Ce3/NAC Before Reaction	599.35
Mn3Ce3/NAC After 150 °C Reaction	547.72
Mn3Ce3/NAC After 200 °C Reaction	571.40

Morphology Evolution

The SEM/EDS images in Figure and Figure , respectively, display morphologies and sizes of the Ce-rich and Mn-rich particles of Mn3Ce3/NAC after HC-SCR reaction testing at 150 and 200 °C. During the experiment, the state of the activated carbon support did not change significantly before and after the reaction. As compared to the as-prepared catalyst, reaction testing caused growth in the sizes of both Mn- and Ce-rich particles, but the EDS data also suggest that the Mn was more highly dispersed after 200 °C testing than after 150 °C testing. This difference may be related to an increased probability of interactions between Ce and Mn which enabled continued, stable NO conversion at 200 °C as opposed to the steady decrease in NO conversion at 150 °C.

Figure 4

Morphology of Mn3Ce3/NAC after reaction testing of 150 °C. (a) SEM image Mn3Ce3/NAC, 2000×. (b) EDS image of Mn/Ce element dispersion.

Figure 5

Morphology of Mn3Ce3/NAC after reaction testing of 200 °C. (a) SEM image Mn3Ce3/NAC, 2003×. (b) EDS image of Mn/Ce element dispersion.

Crystalline Phases

Figure data show the XRD patterns of Mn3Ce3/NAC tested at 200 °C were very similar before and after reaction testing, whereas while the XRD pattern of the catalyst tested at 150 °C was different, the intensity of XRD peaks for the sample after the reaction at 150 °C decreased. Interpreting the exact contributions to peak intensities from the various compounds was complicated, because it required deconvolution of overlapping peaks due amorphous carbon, CeO2, MnO2 and Mn3O4. However, if the intensity contributions from amorphous carbon to the peak at 2θ = 24° and 2θ = 43° were approximately constant—as indicated by the fact that the peak intensity from carbon at 2θ = 43° was mostly unchanged independent of whether the fresh catalysts tested at 150 or 200 °C were examined—then the overall intensity changes in the 2θ = 20–30° composite peak were primarily caused by differences in concentrations and/or crystallinities of the metal oxides. In addition, the intensity of the CeO2 peak at 79.1° was slightly less in the fresh than in catalyst tested at 200 °C, indicating different crystallinity for CeO2 in the catalyst tested at 150 °C; as a consequence, it would contribute to distinct shoulders in peak intensities at 28.5°, 47.5°, and 56.3°. Simultaneously, the peak at 56.3° was most pronounced in the catalysts tested at 200 °C, thereby implying that the concentration and/or crystallinity of MnO2 was distinct in the catalysts tested at 200 °C as compared to the other tested catalysts. Therefore, the MnO2 and CeO2 components of the catalysts tested at 200 °C were more stable than those tested at 150 °C. This stability in the crystalline phase may enable the catalysts tested at 200 °C to maintain the oxidation state of Mn and its catalytic performance.

Figure 6

XRD pattern of Mn3Ce3/NAC catalyst: (a) before reaction testing; (b) after 150 °C reaction testing; (c) after 200 °C reaction testing.

H2-TPR Analysis

The H2-TPR data shown in Figure exhibited a TPR profile that was deconvoluted into three distinct peaks between 200 and 800 °C.[50] Reduction of the weakest Mn–O bonds occurred at the temperature region T1 and was for MnO2 to Mn2O3; the amorphous Mn2O3 was then reduced to Mn3O4 before converting to MnO at temperature T2. The reduction of CeO2 was mainly concentrated between 518  and 787 °C and assigned to the T3 temperature.[51]

Figure 7

H2-TPR profiles of Mn3Ce3/NAC catalysts: (a) before reaction; (b) after 150 °C reaction; (c) after 200 °C reaction.

Before reaction testing, the two main reduction peaks were at 442 and 644 °C, in which the 442 °C peak was attributed to Mn2O3 and the peak at 644 °C was assigned to surface reduction of CeO2; the 442 °C peak could accounted for reduction of surface Ce4+ to Ce3+.[52] For the 150 °C reaction tested sample, the intensity of the 442 °C peak was greater than before testing, indicating more Mn2O3 after testing.[53] From the literature,[54] it is expected that the reduction of MnO to MnO would involve the presence of Mn3O4. However, after 200 °C reaction testing, the higher peak intensity at 644 °C pointed to relatively high concentrations of Ce4+; also, for this sample, the broad, high-intensity reduction peak near 518 °C was due to the reduction of surface oxygen of ceria.[55]

C2H4-TPD/NO-TPD Analysis

The C2H4-TPD and NO-TPD data were shown in Figure . At 80–700 °C, Mn3Ce3/NAC catalysts had 3 desorption peaks for C2H4 and NO. In C2H4-TPD, the desorption peaks of the Mn3Ce3/NAC catalyst were located at 156, 373, and 591 °C. The 156 °C peak corresponded to the medium-strong acid site, and the peaks at 373 and 591 °C were attributed to the strong acid site. It can be seen from the C2H4-TPD spectrum that the distribution of medium-strong acid centers was greater than that of strong acid. The NO-TPD spectrum shows that the desorption peaks of the Mn3Ce3/NAC catalyst were distributed at 160, 316, and 396 °C. At 160 °C, the NO desorption peak was very small; that is, the corresponding medium-strong acid site content was relatively low, which indicated that under low temperature conditions, C2H4 was more easily adsorbed on the surface of Mn3Ce3/NAC catalyst.

Figure 8

C2H4-TPD and NO-TPD of Mn3Ce3/NAC catalysts.

In Situ DRIFTS Study

Figure a shows the DRIFTS spectra of Mn3Ce3/NAC catalysts in a flow condition of NO + O2 + He at 200 °C. When the mixed gas was introduced for 1 min, a strong band appeared at 2350 cm–1, which was attributed to CO2.[56] Over time, other bands at 1910, 1850, 1630, 1590, 1310, and 844 cm–1 were detected. The peak at 1910 cm–1 corresponded to C–H; 1850 cm–1 peak was assigned to acid anhydride; 1630, 1590, and 1310 cm–1 were the characterization signals of NO2–;[57] and the band at 844 cm–1 was attributed to υsC–N. Therefore, nitrates on Mn3Ce3/NAC catalyst were formed from the reaction between NO and O2.

Figure 9

DRIFTS spectra of Mn3Ce3/NAC catalysts in a flow of (a) NO + O2 + He; (b) C2H4 + O2 + He; (c) C2H4 + NO + O2 + He.

Figure b shows the DRIFTS spectra of Mn3Ce3/NAC catalysts in a flow of C2H4 + O2 + He at 200 °C. After the mixed gas was introduced, two strong absorption peaks appeared at 2350 and 2980 cm–1, respectively; the peak at 2350 cm–1 was consistent with the band in Figure , which corresponds to CO2, while the peak at 2980 cm–1 was attributed to the CH2 and CH3 stretching vibrations.[58]

Figure 10

Deconvoluted XPS spectra of the Mn 2p: (a) before reaction; (b) after 150 °C reaction; (c) after 200 °C reaction.

Figure c shows the DRIFTS spectra of Mn3Ce3/NAC catalysts in a flow of C2H4 + NO + O2 + He at 200 °C. At time of 1 min, there were 5 bands, and they were located at 2980, 2350, 984, 949, and 910 cm–1, respectively. Among them, the peak at 2980 cm–1 was assigned to CH2 and CH3 stretching vibrations,[58] the band at 949 cm–1 was attributed to δO–H of R-COOH, the band at 2350 cm–1 was attributed to CO2, the peak at 984 cm–1 was attributed σC–H of olefin, and the band at 910 cm–1 belongs to C–N. The band intensity at 1524 cm–1 was weak, and it belonged to υasCOO–.[56]

Comparing Figure a–c, there was an absorption CO2 peak under the three gas flow conditions, which may be the result of the oxygen-containing functional groups in the activated carbon being oxidized by oxygen. From Figure a, it can be inferred that after the introduction of NO and O2, NO was adsorbed on the surface of the Mn3Ce3/NAC in situ catalyst and generated nitrate. Figure b shows that ethylene had a good adsorption capacity on the Mn3Ce3/NAC catalyst and produced the important intermediate products in the HC-SCR, which can be partially oxidized to generate hydrocarbon activated intermediates. Figure c demonstrated that when C2H4 + NO + O2 was introduced, strong peaks R-COOH and weak peak of unidentate nitrate appeared, which indicated that ethylene was more easily adsorbed on the catalyst surface.

Based on the above results, during the HC-SCR reaction, the C2H4–SCR reaction route was determined as the ethylene being first adsorbed on Mn3Ce3/NAC catalysts to produce the intermediate products containing CH2 or CH3, and then reacted with NO to further produce products containing R-COOH, nitro, and C–N, and finally decomposed into final products H2O, CO2, and N2, which indicated that the C2H4–SCR reaction following the E-R mechanism.[59−61]

Binding State of Constituent Elements

Oxidation states of catalytically active species worked as an essential role in the catalytic performance;[62] hence, XPS analyses of the Mn3Ce3/NAC catalysts were also performed, the data from which are presented in Figures , 11, and 12. As shown in Figure , the main peaks of Mn 2p3/2 deconvoluted into contributions from respective oxidation states, including 642.4, 641.2, and 640.5 eV corresponding to Mn4+, Mn3+, and Mn2+, respectively.[63] Similarly, the peaks at 654.1, 652.8, and 651.1 eV were ascribed to Mn4+, Mn3+, and Mn2+ for Mn 2p1/2, respectively.[63] In addition to contributions to peak intensities from the three oxidation states of Mn, the intensity ratios of Mn4+/(Mn2+ + Mn3+ + Mn4+) and Mn2+/(Mn2+ + Mn3+ + Mn4+)—labeled as Mn4+/Mn and Mn2+/Mn ratios in Table —are presented in Table . The catalyst before and after it was tested at 200 °C had high Mn4+/Mn ratios near 41%, whereas this ratio was 30% after 150 °C reaction testing. Hence, the contribution of Mn4+ was significantly less in the catalyst tested at 150 °C; it matched the XRD results in which the intensity of peaks for the catalyst after the reaction at 150 °C decreased.

Figure 11

Deconvoluted XPS spectra of the Ce 3d: (a) before reaction; (b) after 150 °C reaction; (c) after 200 °C reaction.

Figure 12

Deconvoluted XPS spectra of the O 1s: (a) before reaction; (b) after 150 °C reaction; (c) after 200 °C reaction.

Table 2

Chemical States and the Ratios of Relative Concentrations of Mn, Ce, and O for Different Catalysts

samples	Mn4+/Mnn+ (%)	Mn2+/Mnn+ (%)	Ce4+/(Ce4+ + Ce3+) (%)	Oα (%)	Oβ (%)	Oγ (%)
Before reaction	41.5	4.7	79.8	46.0	43.5	10.5
After 150 °C reaction	30.0	4.5	67.3	50.9	43.8	5.3
After 200 °C reaction	40.9	6.9	81.8	79.0	20.9	0.1

Relative to the contribution of Mn4+ to HC-SCR, it is known that the activity of MnO species follows the order of MnO2 > Mn5O8 > Mn2O3 > Mn3O4 > MnO,[64] suggesting that the existence of high oxidation state Mn was favorable for NO conversion. Furthermore, the acceleration effect of Mn4+ to the oxidation of NO to NO2 has been attested, which could then impact SCR reactions through a “fast SCR” pathway.[65] Less Mn4+ in the catalyst tested at 150 °C was also suggested by the H2-TPR data.

As shown in Figure , the Ce 3d spectra of the fresh and tested catalyst samples were split into nine main peaks. There are four pairs of spin–orbit doublets, and the nine peaks at 882.3, 884.6, 885.8, 888.6, 898.3, 900.8, 904, 907.3, and 916.7 eV were respectively marked as V0, V, V′, V″, U0, U, U′, U″, and U‴. Due to the spin–orbit of the Ce 3d5/2 and Ce 3d3/2, the peaks were labeled V and U, respectively.[66,67] The V0, V, V″, U, U″, and U‴ peaks were associated with the Ce4+, while the peaks denoted as U0, U′, and V′ corresponded to the Ce3+ species. Hence, the coexistence of Ce3+ and Ce4+ in Mn3Ce3/NAC was confirmed through the Ce 3d XPS spectra. It was discovered that the coexisting ion pair Ce3+/Ce4+ species could create a charge imbalance, unsaturated chemical bonds, and oxygen vacancies on surfaces; these were in favor of the formation of chemisorbed oxygen during reaction testing.[68]

As shown in Table , the ratio of [Ce4+/(Ce4+ + Ce3+)] in the fresh catalyst and after 200 °C reaction testing was about 80%. According to the literature,[69] the incorporation of Ce in Mn compounds can create a Mn–O–Ce bridge during CeO2 formation that also strengthens the bond between Mn and oxygen and promotes the valence of Mn from Mn3+ to Mn4+ through Mn2O3 + 2CeO2 → 2MnO2 + Ce2O3. Hence, the existence of an oxygen vacancy and facile Ce3+/Ce4+ couple could prevent the reduction of MnO2 to Mn2O3 while simultaneously creating larger Ce4+ concentrations through the redox cycle of Mn3+ + Ce4+ ↔ Mn4+ + Ce3+.[66] Then, the presence of a stable Mn4+ is suggested to facilitate HC-SCR reactions. In other words, the 79% and 81% Ce4+ ratios on the fresh and 200 °C reaction tested samples, and the stable NO conversion, are in accordance with this interpretation.

Figure describes the XPS spectra for O 1s of the catalysts in which the deconvolution of the peaks was accomplished using three peaks. The binding energies of peaks were approximately 529.6, 531.1, and 533.2 eV, credited to lattice oxygen (O2–, marked as Oα), surface chemical oxygen (O22– or O–, marked as Oβ), and the oxygen species in hydroxide groups which was marked as Oγ, respectively.[66] The relative intensity of Oβ changed intensely after the HC-SCR reactions. Thus, the surface chemisorbed oxygen has also been more correlated to higher catalytic activity than lattice oxygen as a result of its higher mobility.[69] As shown in Table , the ratios of Oβ were greater than for most other catalysts,[70] the ratio of Oβ/(Oα + Oβ) on the Mn3Ce3/NAC catalysts before reaction, after 150 °C reaction, and after 200 °C reaction are 0.49, 0.46, and 0.21; for the SCR reaction, the high ratio of Oβ/(Oα + Oβ) means that NO is more easily oxidized to NO2, which is an important intermediate to generate N2, and suggests ample presence of facile oxygen for catalytic reactivity.

Therefore, the analytical characterization data of Mn3Ce3/NAC catalysts before and after the reaction testing point to a reaction sequence during HC-SCR as depicted in Figure . The presence of CeO2 in a sufficient concentration is critical for stable NO conversion, because it helps to prevent the reduction of MnO2 to Mn2O3 by promoting the forward reaction and simultaneously, through synergistic interactions, promotes the establishment of relatively high concentrations of facile, absorbed oxygen on the catalyst surface. Therefore, C2H4 is more easily oxidized to produce the hydrocarbon activated intermediate, −COOH, which promotes the C2H4–SCR reaction. When the CeO2 concentrations were not high enough or the reaction temperature was too low, a precipitous decrease in NO activity occurred immediately after reaction testing had begun, because the Ce species were agglomerated into small, isolated CeO2 crystallites which did not effectively interact with the MnO and did not mitigate the formation of low-activity Mn3O4 species.

Figure 13

Synergistic mechanism for HC-SCR of NO on Mn3Ce3/NAC catalysts.

Experimental Methods

Catalyst Preparation

Active carbon used was purchased from MACKLIN with particle size range of 1000–2350 μm. The raw AC was first treated with 10% HNO3 solution for 4 h under room temperature and washed with deionized water until the pH of the wastewater became neutral, and finally the AC was dried under air atmosphere at 140 °C for 14 h. These HNO3-treated AC supports were named as NAC.

The Mn–Ce/AC catalysts with MnO2 and CeO2 loadings were prepared by impregnation of the NAC using Mn(NO3)2·4H2O and Ce(NO3)3·6H2O solutions to attain a Mn loading of about 3.0 wt % and Ce loadings of 3.0, 5.0, and 7.0 wt %; sonication of these mixtures was at room temperature for 2 h. Then, the hybrids stand for 12 h, after which they were filtered and dried in a vacuum oven heated to 110 °C, and finally calcined at 400 °C for 2 h in a hermetic muffle furnace under N2 atmosphere. The synthesized materials were labeled as Mn3Ce/NAC, where X = 3, 5, 7.

Catalyst Characterization

The synthesized materials were characterized analytically before and after reaction testing. After degassing overnight at 160 °C and isothermal N2 adsorption–desorption measurements taken at 77 K, surface area and porosity were measured in a Micromeritics ASAP 2020 analyzer.

Crystalline structures were performed by X-ray diffraction (XRD) using Cu Kα irradiation with a 2θ range of 90°. The microstructure was examined by scanning electron microscope (SEM) with energy-dispersive X-ray spectroscopy (EDS). X-ray photoelectron spectroscopy (XPS) was performed on a Thermo Fisher ESCALAB 250Xi instrument, and the data were calibrated using the ubiquitous C 1s peak at 284.8 eV.

The temperature-programmed reduction (TPR) data were acquired using an AutoChem II 2920 instrument; approximately 0.1 g samples were placed in a U-shaped quartz tube which was then pretreated in O2 flow at 300 °C for 1 h. After the samples were cooled to room temperature, 5% H2 in helium was flowed through the tube at a flow rate of 30 mL/min while the temperature was increased from room temperature to 900 °C at a rate of 10 °C/min.

The temperature-programmed desorption (TPD) study was performed on an AutoChem II 2920 instrument: 0.1 g sample was pretreated in helium at 300 °C for 1 h; afterward the sample was cooled to room temperature, and C2H4 or NO in helium was purged to the tube at a flow rate of 30 mL/min as the temperature was increased from 80 to 700 °C at a rate of 10 °C/min.

In situ DRIFTS was performed on a VERTEX 80 V instrument: 0.2 g sample was pretreated in helium at 200 °C for 30 min, and signals were collected as the background spectrum. Then, the NO + O2, C2H4 + O2, and C2H4 + NO + O2 was separately flowed into the reactor at 30 mL/min, and the sample spectra at 1, 3, 5, 10, 15, 20, 25, and 30 min were collected (scanning times, 128; resolution, 4 cm–1, scanning range, 4000–600 cm–1).

Catalytic Activity Testing

The activity evaluation equipment of the prepared catalysts (Figure ) involved a fixed bed, a gas analyzer (MRU MGA5), a gas cleaning unit, several mass flow controllers (MFCs, MF SHY 400), and a quartz tube reactor (I.D. = 10 mm) heated by a temperature-controlled furnace; the initial weight of a sample for each test was 10 g. Total gas flow at a rate of 1500 mL/min was controlled by mass flow controllers (MFCs, MF SHY 400). The components of the simulated flue gas were 500 ppm of NO, 250 ppm of C2H4, 3 vol % O2, and N2 as a balance. The catalyst sample was in N2 atmosphere during the heating procedure; once the temperature remained unchanged, the mixture was flowed into the reactor, and then NO and NO gas concentrations at the outlet of the reactor were analyzed using a flue gas analyzer (MRU). From these concentration data, NO conversion and N2 selectivity data were calculated through the following equation:

Figure 14

Schematic of the bench-scale setup.

Conclusions

This study focused on examining the potential development of stable and active Mn3Ce/NAC catalysts used for HC-SCR with C2H4 as a reductant and on understanding catalyst physical and chemical properties important for NO conversion with temperatures at or below 200 °C. The catalysts were prepared, tested for HC-SCR catalytic activity, and analyzed before and after reaction testing using surface sensitive and bulk techniques. The conclusions include the following:

The addition of CeO2 to the MnO established stable and high HC-SCR catalytic performance when the reaction temperature was at least 200 °C and the CeO/MnO concentration ratio was 1 or reaction temperature was below 200 °C and the CeO/MnO concentration ratio was > 1.

The reaction mechanism followed the E-R mechanism, and ethylene was first adsorbed on the Mn3Ce3/NAC catalysts to produce the intermediate products containing CH2 or CH3, which can be partially oxidized to generate hydrocarbon activated intermediate, R-COOH.

Stable activity was proposed to be a result of the formation of CeO2 at a sufficient concentration that could prevent the reduction of Mn4+ and, simultaneously, the creation of Ce3+/Ce4+ redox couples that would help to reoxidize any reduced Mn.

The addition of CeO2 promoted the formation of Mn–O–Ce bridges with MnO that then helped to enhance the concentration of surface chemisorbed oxygen and oxygen vacancies. This study provided new and intriguing insight into approaches to instill effective catalytic activity into transition metal oxide catalysts for low-temperature HC-SCR applications.