Effect of Indium Addition on the Low-Temperature Selective Catalytic Reduction of NO x by NH3 over MnCeO x Catalysts: The Promotion Effect and Mechanism.

A MnCeInO x catalyst was prepared by a coprecipitation method for denitrification of NH3-SCR (selective catalytic reduction). The catalysts were characterized by Fourier transform infrared spectroscopy (FTIR), X-ray diffractometry, scanning electron microscopy, X-ray photoelectron spectroscopy, Brunauer-Emmett-Teller analysis, H2 temperature-programmed reduction, and NH3 temperature-programmed desorption. The NH3-SCR activity and H2O and SO2 resistance of the catalysts were evaluated. The test results showed that the SCR and water resistance and sulfur resistance were good in the range of 125-225 °C. The calcination temperature of the Mn6Ce0.3In0.7O x catalyst preparation was studied. The crystallization of the Mn6Ce0.3In0.7O x catalyst was poor when calcined at 300 °C; however, the crystallization is excessive at a 500 °C calcination temperature. The influence of space velocity on the performance of the catalyst is great at 100-225 °C. FTIR test results showed that indium distribution on the surface of the catalyst reduced the content of sulfate on the surface, protected the acidic site of MnCe, and improved the sulfur resistance of the catalyst. The excellent performance of the Mn6Ce0.3In0.7O x catalyst may be due to its high content of Mn4+, surface adsorbed oxygen species, high specific surface area, redox sites and acid sites on the surface, high turnover frequency, and low apparent activation energy.

Highlights

A novel NH3-SCR catalyst was developed by doping indium into MnCeO.

The MnCeInO catalyst showed >90% conversion of NO at 125–225 °C.

The resistance to H2O and SO2 of the Mn6Ce0.3In0.7O catalyst was enhanced significantly.

The mechanism of catalysis of NH3-SCR of the novel catalyst was analyzed.

The doping of indium improved the TOF value of the catalyst and reduced the apparent activation energy of the catalyst.

Introduction

NO is a harmful pollutant produced by fossil fuels mainly from industrial processes and residential life. The natural degradation way is to combine it with water to produce nitric acid into the earth with rain. A large amount of nitric acid combined with rain will form acid rain, and gaseous NO will form photochemical smog, which has a great impact on human health and ecological balance.[1] In recent years, the government had higher and higher requirements for NO emission standards of enterprises.[2] Many denitrification technologies have been developed to meet the standards.[3−5] At present, the vanadium catalyst (V2O5-WO3/TiO2) is widely used in industry. Due to its strict high-temperature operating window (290–400 °C) and toxic pollution of vanadium sublimation at high temperatures, the working environment of the catalyst has many restrictions.[6,7] With the improvement of waste heat utilization technology of flue gas in power plants, the requirement of flue gas denitration for catalyst temperature is higher and higher. Therefore, the development of a low-temperature denitration catalyst with high efficiency is very promising.[8,9] NH3-SCR (selective catalytic reduction) is an approach for effective NO reduction.[10] Therefore, it is very promising to develop catalysts suitable for low-temperature NH3-SCR technology instead of high-temperature active catalysts.

Among all the studied metals, MnO is widely studied for its excellent low-temperature catalytic activity, but it cannot be applied in practice due to its poor water and sulfur resistance.[11] Other metal oxides have been added to further improve their low-temperature activity and stability. Because of its excellent redox performance, cerium was used as an enhancer to dope the original catalyst to improve the catalyst activity.[12] Li et al. prepared Mn–CeO nanospheres using the geothermal approach with excellent denitrification performance but only at low space velocity.[13] Andreoli et al. prepared a MnO-CeO catalyst through the amino acetic acid method.[14] Although it has an excellent conversion rate at 150–280 °C, its selectivity is poor. Although cerium has improved the sulfur resistance of the MnO catalyst, it still needs to be further improved.[15−17] Decolatti et al. synthesized In-NH4-zeolites and found that indium species could promote the oxidation of NO to NO2. The higher the indium content meant the better the denitrification effect.[18] Pan et al. found that surface indium species can react with SO2 and H2O to produce In2(SO4)3.[19] It was found that the low-temperature denitrification performance could be significantly improved by H2 treatment at 400 °C for 60 min, although the sulfate radical could not be completely removed.

In this work, the denitrification performance of MnCeO and MnCeInO catalysts with different ratios of In and the stability in the presence of water vapor and SO2 have been tested using an NH3-SCR denitrification device at 100–275 °C. The gas hourly space velocity (GHSV) was 120,000 h–1. The catalysts were characterized by FTIR, SEM, XRD, BET, XPS, H2-TPR, and NH3-TPD. The possible reaction pathways were studied.

Results and Discussion

Catalyst Performance Analysis

Figure shows the NO conversion efficiency of MnCeO and a series of MnCeInO with different proportions. The T80 (NO conversion efficiency of >80%) of Mn6CeO is at 125–200 °C, and the highest conversion rate is 91.2%. When Ce:In = 3:7, the catalyst shows the best conversion efficiency of T90 in the range of 125–225 °C and the best conversion rate is 94.5%. This not only enlarges the Mn6CeO temperature operating window but also improves the conversion efficiency. It can also be seen from Figure B that compared with the Mn6CeO catalyst, doping of indium also improves the selectivity of N2. Figure C shows the effect of different calcination temperatures and different space velocities for the NO conversion efficiency with the Mn6Ce0.3In0.7O catalyst. As shown by the figure, when the space velocity is too high, the conversion efficiency of the catalyst decreases significantly. When the space velocity decreases, the catalyst conversion rate increases further in the range of 100–175 °C, indicating that increasing the volume of the catalysts could increase the number of active sites and facilitate the processing of more feed gas. The calcination temperature has a great influence on the conversion efficiency of the catalyst. At 300 °C, the catalyst showed poor catalytic activity, the best conversion rate was only 90.8%, and T80 had a temperature operating window of 125–250 °C. When the calcination temperature is 500 °C, the conversion efficiency of the catalyst is slightly lower than the calcination temperature of 400 °C but its T90 still has a wide temperature operating window of 125–225 °C. It can be seen from the XRD diagram that the crystallization of the catalyst after calcination at 500 °C is higher than the other two calcination temperatures, which leads to the agglomeration of CeO2 at the catalyst surface and the generation of more Mn2O3, affecting the catalytic efficiency. XRD patterns at a calcination temperature of 300 °C showed wide and mixed peaks, indicating that the catalyst was poorly formed and not fully calcined.

Figure 1

(A) NO conversion. (B) N2 selectivity. Reaction conditions: 500 ppm NH3, 500 ppm NO, 5 vol % O2, and balance N2, with a GHSV of 120,000 h–1. (C) Different temperatures and GHSVs over different catalysts.

Figure A shows the water and sulfur resistance test results of Mn6Ce0.3In0.7O and Mn6CeO catalysts at 150 °C. First of all, the catalyst was stably exposed to raw gas at 150 °C for 1 h, 5% H2O was passed through the catalyst for continuous testing for 5 h, and then, the water vapor was closed. Then, 100 ppm SO2 was passed one hour after catalyst recovery, SO2 was closed after the continuous test for 5 h, and then, we waited for recovery for another hour. On this basis, 5% H2O and 100 ppm SO2 were added simultaneously, water vapor and SO2 were closed after 5 h of monitoring, and the recovery of the catalytic efficiency was detected for 1 h.

Figure 2

(A) Water and sulfur resistance of Mn6Ce0.3In0.7O and Mn6CeO catalysts. (B) FTIR spectra of Mn6Ce0.3In0.7O and Mn6CeO catalysts treated in simulated flue gas at 150 °C for 60 min. Reaction conditions: 500 ppm NH3, 500 ppm NO, 5 vol % O2, 5 vol % H2O, 100 ppm SO2, and balance N2 (GHSV = 120,000 h–1).

In practical engineering applications, H2O and SO2 are the influencing factors that cannot be ignored. The sulfurization of catalysts and the formation of NH4HSO4 are the key reasons leading to the reduction of catalytic efficiency of catalysts.[22] In Figure A, the water resistance of the two catalysts is similar, and the conversion rate decreases by 8%. After H2O is closed, Mn6Ce0.3In0.7O shows a stronger recovery ability than the Mn6CeO catalyst, and the conversion efficiency of Mn6Ce0.3In0.7O returns to the state before water is added. The Mn6CeO catalyst recovered 5%. When 100 ppm SO2 was introduced, the catalytic efficiency of Mn6Ce0.3In0.7O suddenly dropped, which may be due to part of the active sites on the catalyst surface being covered by sulfide or ammonium sulfate when SO2 was passed through the catalyst in the early stage. After 1 h of SO2 induction, the catalytic efficiency recovered 91.6%, which was related to the decomposition of ammonium sulfate on the catalyst surface, releasing part of the active sites and causing the efficiency of the catalyst to rebound. Then, the conversion rate gradually decreased to 89.4% with the increase in time and recovered 93% after the closure of SO2. The catalytic efficiency of Mn6CeO decreased gradually with the increase in time, decreased to 76% after 5 h, and recovered 81.5% after closing SO2. When the water and sulfur resistance of the two catalysts was tested, Mn6Ce0.3In0.7O showed a better effect than Mn6CeO. The conversion efficiency of Mn6Ce0.3In0.7O and Mn6CeO decreased to 82.4 and 67% after the 5 h test, respectively. The conversion efficiency of Mn6Ce0.3In0.7O and Mn6CeO recovered 87.2 and 70.7% after water vapor and SO2 were closed, respectively. The addition of indium significantly improved the resistance ability of the Mn6CeO catalyst to H2O and SO2. According to the recovery ability of the Mn6Ce0.3In0.7O catalyst after the test of sulfur resistance, SO2 had little effect on it, and the main effect was from water vapor. The influence on the recovery ability of the two catalysts through the H2O and SO2 resistance tests revealed that a part of the active site of the catalyst may be covered by (NH4)2SO4.

To obtain the difference of the surface structure of the catalysts after the water and sulfur resistance test, the catalysts were characterized by Fourier transform infrared spectroscopy (FTIR). Figure B shows the FTIR spectra of Mn6Ce0.3In0.7O and Mn6CeO catalysts after H2O and SO2 resistance tests. In FTIR tests of the Mn6CeO catalyst, 3414 cm–1 was attributed to the vibration of the O–H bond of H2O.[23,24] In the FTIR test of the Mn6Ce0.3In0.7O catalyst, 3390 cm–1 was attributed to the N–H tensile vibration mode of NH3.[24,25] According to literature reports, 3200 cm–1 was attributed to the formation with other forms of NH3 and NH4+.[24] The 1631–1627 cm–1 peak belongs to adsorbed NO2.[26] The 1450–1443 cm–1 peak belongs to NH4+ formed by adsorption at the Brønsted acidic sites.[27,28] The 1400–1383 cm–1 peak was attributed to ammonium sulfate[29] and 1114–1070 cm–1 to sulfate.[30] The 1265 cm–1 peak was attributed to monotonic nitrite.[31] The 863–862 cm–1 peak belongs to physical adsorption or weak phase adsorption of NH3. In the wavelength range of less than 800 cm–1, it belongs to the vibration between metal and oxygen atoms, and 558–569 cm–1 was related to the vibration of the Mn–O bond.[32] By comparing the results of the two catalysts, the two catalysts both contain single-toothed nitrate, adsorbed NO2, adsorbed NH3, and NH4+ but do not contain double-toothed nitrate. In addition, they all contain ammonium sulfate and sulfate. The results showed that SO2 inhibited the formation of nitrates on the Mn6CeO catalyst and had little effect on the adsorption of nitrates and ammonia on the monotone. It is not difficult to see that the peak intensity of ammonium sulfate and sulfate on the Mn6Ce0.3In0.7O catalyst decreases obviously in the FTIR diagram, and the peak intensity increases in the range of 3000–3750 cm–1, which indicates that indium doping could effectively reduce the formation of ammonium sulfate and sulfate at the catalyst. Moreover, it can improve the chemisorption of NH3 and the formation of NH4+.

XRD Analysis

Figure A presents the XRD results for the MnCeO and MnCeInO series. According to JADE 6 software, the Mn6CeO sample contains sharp XRD peaks of the MnO2 phase (PDF no. 89-5171) and the CeO2 phase (PDF no. 34-0394). The presence of MnO2, CeO2, In2O3, and Mn2O3 (PDF no. 24-0508) phases can be observed in the MnCeInO series. With the decrease in the cerium content and the increase in the indium content, the XRD peak of the CeO2 phase becomes weaker, and the peak strength of 37.2° of the (100) crystal of MnO2 gets stronger. With the increase in the indium content, the (102) plane diffraction peak at 56.47° of MnO2 decreases gradually, indicating that MnO exists as an amorphous component.[33] The results indicated that indium was doped to inhibit the crystallization of CeO2 and promote the growth of the MnO2(100) crystal. At the same time, the In2O3 phase (PDF no. 22-0336) gradually appeared, indicating that the In2O3 structure formed on the surface of MnCeO.

Figure 3

(A) XRD results of MnCeO and MnCeInO series. (B) XRD results of the Mn6Ce0.3In0.7O catalyst at different calcination temperatures.

In Figure B, the calcination temperatures of Mn6Ce0.3In0.7O catalysts were compared. The catalysts calcined at 300 °C have obvious diffraction peaks between 30 and 45° but no diffraction peaks in other ranges. It indicates that the catalyst is not fully formed. The catalyst calcined at 500 °C has an obvious diffraction peak of CeO2 at 2θ = 28°, which indicates that the dispersed cerium can be condensed by calcination at high temperatures. At 2θ = 33°, the Mn2O3 crystal peak is formed, indicating that more Mn2O3 is produced on the surface of the catalyst by calcination at high temperatures. The agglomeration of CeO2 and more Mn3+ negatively affects the performance of the catalyst. Combined with the denitration test results, studies showed that the catalyst calcined at 300 °C had the worst NO conversion, the efficiency was slightly better at 500 °C, and the efficiency was the best at 400 °C. In summary, the preparation temperature of calcination at 400 °C is the best preparation temperature for the catalyst.

SEM Result Analysis

Figure shows the SEM morphology of (a–c) Mn6Ce0.3In0.7O and (d–f) Mn6CeO catalysts. The comparison of Figure a and Figure d shows that the Mn6CeO catalyst with more cerium content has many CeO2 flakes dispersed around spherical MnO. Mn6Ce0.3In0.7O is spherical in various sizes and contains a small amount of CeO2 flakes. According to the comparison of Figure c and Figure f, the particle diameter of the Mn6Ce0.3In0.7O catalyst is 7.5 μm, and the surface of the Mn6Ce0.3In0.7O catalyst is uneven and covered like microvilli. The particle diameter of the Mn6CeO catalyst is 12 μm, and the surface appears smooth and flat. The addition of indium changed the appearance and size of the Mn6CeO catalyst, which increased the specific surface area of the Mn6Ce0.3In0.7O catalyst.

Figure 4

SEM morphology of (a–c) Mn6Ce0.3In0.7O and (d–f) Mn6CeO catalysts.

Variation of the Specific Surface Area and the Pore Structure

Importantly, a larger specific surface area can provide more catalytic active sites and reaction paths for SCR catalytic reactions. The SEM results indicated that the morphology of the two catalysts also changed, so the difference between them was quantitatively studied by the N2 adsorption/desorption isothermal curve and pore size distribution. Figure presents the nitrogen adsorption/desorption isotherm and pore size distribution of Mn6Ce0.3In0.7O and Mn6CeO catalysts. Both catalysts exhibit a typical type IV isotherm, indicating that both materials are mesoporous. By comparing the hysteresis loop, the two catalysts showed obvious differences. The Mn6CeO catalyst exhibits a typical H4-type hysteresis loop and has significant adsorption capacity at the low end of P/P0, indicating that its pore structure is narrow and fractured.[34] The Mn6Ce0.3In0.7O catalyst exhibits an isotherm type of the H3 hysteresis loop at a high relative pressure between 0.8 and 0.99, indicating that the surface of the catalyst has a mesoporous structure with flat slit channels. The difference can also be seen by considering the pore size distribution in Figure . The pore size distribution of the Mn6Ce0.3In0.7O catalyst is between 2 and 20 nm, and the pore volume distribution is between 0 and 0.23 cm3/g·nm. The pore size and volume distributions of the Mn6CeO catalyst are 3–18 nm and 0–0.15 cm3/g·nm, respectively. Table lists the specific surface area, pore volume, and pore size of Mn6Ce0.3In0.7O and Mn6CeO catalysts. Although the pore size of the catalyst decreases with indium doping, the specific surface area increases by 23.7%. The results are consistent with those of the hysteresis loop and SEM. Therefore, the doping of indium increases the pore volume of the catalyst, increases the surface area of the catalyst, and provides more catalytic active sites.

Figure 5

N2 physisorption isotherms and pore size distribution of Mn6Ce0.3In0.7O and Mn6CeO catalysts.

Figure 6

XPS Mn 2p (A), O 1s (B), Ce 3d (C), and In 3d (D) spectra of Mn6CeO and Mn6Ce0.3In0.7O catalysts.

Table 1

BET Specific Surface Area and Pore Characterization of the Samples

samples	BET surface area (m2/g)	pore volume (cm3/g)	pore diameter (nm)
Mn6CeOx	113.7	0.15	3.9
Mn6Ce0.3In0.7Ox	140.7	0.23	3.7

Analysis of Surface Element Valence States

In order to understand the properties of elements on the catalyst surface, XPS tests were carried out on Mn6Ce0.3In0.7O and Mn6CeO catalysts, as shown in Figure . The Mn 2p, O 1s, Ce 3d, and In 3d spectra of the catalyst were fitted by XPSPEAK4.1. Figure A shows the deconvoluted peaks of Mn 2p for catalysts Mn6Ce0.3In0.7O and Mn6CeO. The XPS spectra of the Mn 2p region showed a pair of peaks for all the samples, which were attributed to Mn 2p3/2 and Mn 2p1/2, respectively. Three peaks of Mn 2p3/2 of the two catalysts can be observed (Figure A): Mn2+ (640.9–640.8 eV), Mn3+ (642.0 eV), and Mn4+ (643.4–643.3 eV).[34−36] Mn ions in Mn6Ce0.3In0.7O have a variety of valence states, which makes it easy to form a redox pair of Mn/Mn(, causing a good NH3-SCR activity. The relative proportion of Mn4+ in the indium-doped Mn6Ce0.3In0.7O catalyst increased compared to Mn6CeO (Table ). Corresponding to the XRD results, the increase in the high valence state of Mn enhances the oxidation capacity of the Mn6Ce0.3In0.7O catalyst, thus improving the SCR performance.

Table 2

Surface Element Valence States and Relative Contents of Mn, O, and Ce

samples	Mn2+/Mn	Mn3+/Mn	Mn4+/Mn	Oα/O	Oβ/O	Ce3+/Ce	Ce4+/Ce
Mn6CeOx	31.5%	41.1%	27.4%	47.2%	52.8%	18.3%	81.7%
Mn6Ce0.3In0.7Ox	32.3%	36.4%	31.3%	54.7%	45.3%	21.1%	78.9%

Figure B shows the deconvoluted XPS spectra of O 1s in Mn6CeO and Mn6Ce0.3In0.7O catalysts. The peak at 529.49 eV corresponds to the lattice oxygen O2– (Oβ). The peaks at 530.5–531.3 eV correspond to surface oxygen (Oα), such as groups belonging to defective oxides and hydroxyl oxygen (e.g., O22– or O–).[37−39] Since Oα is more reactive and migrates more easily than Oβ, it could contribute to the oxidation of NO to NO2 in the SCR reaction, allowing the catalyst to exhibit a better performance in the oxidation reaction.[40−42] The calculated Oα ratio of the Mn6Ce0.3In0.7O catalyst is shown in Table . The surface oxygen Oα content of the Mn6Ce0.3In0.7O catalyst (54.71%) is higher than that of Mn6CeO (47.17%), indicating that the Mn6Ce0.3In0.7O catalyst has more surface oxygen Oα than the Mn6CeO catalyst, which is beneficial for fast SCR reactions (4NH3 + 2NO + 2NO2 → 4 N2 + 6H2O) at low temperatures.

Figure C shows the fitting spectrum of Ce 3d. The XPS spectra of the Ce 3d region showed a pair of peaks for all the samples, which were attributed to Ce 3d5/2 and Ce 3d3/2, denoted by “u” and “v” corresponding to the spin–orbit component of Ce 3d5/2 and Ce 3d3/2, respectively. It can be observed that there are 8 peaks of u, u1, u2, u3, v, v1, v2, and v3 for Ce 3d where v1 and u1 are Ce3+ and the other peaks are Ce4+.[29,43,44] On the whole, the cerium element in the two catalysts mainly exists in the form of Ce4+ ions, and only a small amount of Ce3+ exists. In Table , the Ce3+/Ce ratios of Mn6Ce0.3In0.7O and Mn6CeO catalysts are 21.1 and 18.3%, respectively. It indicates that the doping of indium leads to the decrease in the relative content of Ce4+, which is because the In3+ ions inserted into the catalyst replace the position of Ce4+, thus changing the oxygen content in the sintering process of the catalyst, transforming Ce4+ into Ce3+.[45] Increasing the concentration of Ce3+ at the catalyst surface can promote the formation of charge imbalance, unsaturated bonds, and vacancies.[33,46] Therefore, the following processes may occur between Ce4+ and Ce3+ on the catalyst surface in the SCR reaction: (1) 2CeO2 → Ce2O3+ O* and (2) Ce2O3 + 1/2O2 → 2CeO2.[29,47] In this process, the oxygen in the flue gas is adsorbed and dissociated on the catalyst surface through the oxygen vacancy to produce oxygen with high fluidity and promote the oxidation of NO to NO2.

Figure D shows the deconvolution of the In 3d5/2 signal, and two peaks can be obtained, namely, a binding energy of 444.34 eV(structure related to In2O3 species) and a binding energy of 444.91 eV (structure similar to InO+ species).[48,49] Surface (InO)+ is considered to be the active site in the SCR reaction. The active site can bind to gaseous NH3 to form adsorbed ammonia and then dissociate NH3 to form −NH2 and [In(OH)],[50] and the formation of these groups contributes to the fast SCR reaction.

Redox Property

The redox capacity of the catalyst was characterized by H2 temperature-programmed reduction (H2-TPR). It is shown in Figure . After the split peak, it was observed that there are four reduction peaks for Mn6CeO and Mn6Ce0.3In0.7O catalysts. By comparing the H2 reduction curves of the two catalysts, the peaks of Mn6CeO in the range of 100–300 °C can be divided into three peaks centered at 243, 265, and 394 °C. These three reduction peaks correspond to the reduction of MnO2 to nonstoichiometric dispersed MnO (1.5 < x < 2). At this point, the exposed high flow of oxygen on the catalyst surface was removed. MnO is then reduced to Mn2O3, at which time part of the lattice oxygen on the catalyst surface decreases. Finally, Mn2O3 is reduced to Mn3O4 and further to MnO.[51−53] Chen et al.[54] reported that the surface Ce4+ to Ce3+ reduction process also occurred at 394 °C. A very low and flat peak was observed at the center of 735 °C. According to the literature, the peak in the range of temperature more than 700 °C corresponds to the reduction of the surface and bulk of cerium oxide.[55]

Figure 7

H2-TPR profiles of catalysts.

Compared with the Mn6CeO catalyst, the Mn6Ce0.3In0.7O catalyst exhibited a similar reduction peak in the temperature range of less than 400 °C. However, the peak values of these reduction peaks all shifted to the direction of low temperatures. The results indicate that the surface of the Mn6CeO catalyst doped with indium is more prone to electron transfer, indicating that the catalyst has a lower SCR activity temperature, thus improving the catalytic performance. Peaks greater than 700 °C were not observed, possibly due to a decrease in the cerium content. There was no reduction peak, or the reduction peak moved to a higher-temperature region, which to some extent reduced the oxidation of cerium in the catalyst, inhibited the formation of nitrous oxide, and reduced the nonselective catalytic oxidation of NH3. This may be one of the reasons for increased N2 selectivity. It can be seen that a new reduction peak is added at 458 °C, which is considered to be the indium phase prone to surface reduction, such as (InO)+ and InOy,[19,56] as shown in the following formula:

After indium doping, indium enters into the lattice of MnO2, and the interaction between the components of the catalyst promotes the reduction of manganite and indium ions, improving the catalytic performance of the SCR catalyst. The reported H2 unit consumption of the catalyst Mn6CeO (7.41 μmol·g–1) was less than that of Mn6Ce0.3In0.7O (8.62 μmol·g–1). The results showed that the reduction peak positions of the two catalysts were not significantly different, but the reduction peaks shifted to low temperatures, indicating that indium doping improved the reduction properties of the Mn–Ce oxides to some extent, which was beneficial to the denitrification performance of the catalyst.

Acidity of the Catalysts

The acidity of Mn6CeO and Mn6Ce0.3In0.7O catalysts was studied by NH3-TPD experiments, as shown in Figure . The amount and intensity of acid sites at the catalyst surface and the activation of NH3 at the catalyst surface were measured by the NH3-TPD technique. In the temperature range of 50–800 °C, Origin 2021 software Gaussian fitting was used to obtain six analytical peaks for catalysts Mn6CeO and Mn6Ce0.3In0.7O. These peaks consisted of peak 1 (100 and 106 °C), peak 2 (147 and 145 °C), peak 3 (230 and 223 °C), peak 4 (442 and 458 °C), peak 5 (525 and 492 °C), and peak 6 (580 and 548 °C). According to literature reports, below 350 °C is a weak acid site, and above 350 °C is a medium-strong acid site.[57,58] Peaks 1, 2, and 3 belong to weak acid sites, which are composed of physical or weak chemical adsorption of NH3 and NH4+ produced by the combination of NH3 and surface hydroxyl groups. Peaks 4, 5, and 6 belong to medium-strong acid sites and are caused by the adsorption of NH3 at the Brønsted or Lewis acid sites. It is well-known that the desorption peak position and the desorption peak area of a catalyst correspond to the strength and quantity of the acid, respectively.[59] The amount of surface acid sites of Mn6CeO and Mn6Ce0.3In0.7O catalysts can be estimated after integrating the NH3-TPD curves. The integral area shown in Figure indicated that the acid content of the Mn6CeO catalyst was 18.7 μmol·g–1, which was less than that of the Mn6Ce0.3In0.7O catalyst (19.3 μmol·g–1). The results showed that the doping of indium in Mn6CeO catalysts could improve the number of surface acid sites and enhance the adsorption of NH3 on the catalysts to a certain extent.

Figure 8

NH3-TPD profiles of Mn6CeO and Mn6Ce0.3In0.7O catalysts.

Kinetic Study

Arrhenius plots of Mn6Ce0.3In0.7O and Mn6CeO catalysts in the range of 110–140 °C are shown in Figure A. The turnover frequency (TOF) of Mn6Ce0.3In0.7O and Mn6CeO catalysts in the range of 100–220 °C is shown in Figure B. The test method was to fully mix 0.01 g of the catalyst and 0.09 g of SiO2 for testing in an NH3-SCR evaluation device to ensure that the reactivity was not affected by diffusion.[21,60] It can be seen from Figure A that the apparent activation energy (Ea) of the Mn6Ce0.3In0.7O catalyst (49.8 kJ /mol) is less than that of Mn6CeO (78.4 kJ /mol), indicating that it is easier for the Mn6Ce0.3In0.7O catalyst to promote the reaction than Mn6CeO. In Figure B, the TOF value of the Mn6Ce0.3In0.7O catalyst increased as the temperature increased. The higher the temperature value, the higher the TOF value and the better the SCR activity, which finally tended to balance after 200 °C. The TOF value of the Mn6CeO catalyst also increased with the increase in temperature and reached the maximum value at 190 °C and then gradually decreased. This is the same as its SCR activity.

Figure 9

(A) Arrhenius plots of Mn6Ce0.3In0.7O and Mn6CeO catalysts. (B) TOF over Mn6Ce0.3In0.7O and Mn6CeO catalysts at different temperatures.

Discussion

The results show that the MnCeO catalyst has a temperature operating window of T80 of 125–200 °C at GHSV = 120,000 h–1 and shows good activity but poor activity in other temperature test sections. The activation of the MnCeO catalyst was promoted by adding indium. For example, the temperature window of the catalyst T90 increases from 150–175 to 125–225 °C. The activation energy of the Mn6Ce0.3In0.7O catalyst is lower than that of the Mn6CeO catalyst obtained from the Arrhenius plot (Figure A) results. In the range of 100–220 °C, the TOF value calculated based on NH3-TPD data (Figure B) is larger for the Mn6Ce0.3In0.7O catalyst. The results showed that the doping of indium could effectively improve the synergistic interaction between active sites, increase the TOF value of the Mn6CeO catalyst, and improve the catalytic performance of the catalyst. Mn6Ce0.3In0.7O and Mn6CeO catalysts have been compared in water and sulfur resistance tests (Figure A). It can be found that both catalysts show a good recovery ability in water resistance tests. In addition, Mn6Ce0.3In0.7O showed a better sulfur resistance than Mn6Ce0.3In0.7O, and neither of the two catalysts returned to the state before adding SO2. The Mn6Ce0.3In0.7O catalyst has more obvious advantages when both H2O and SO2 are present and has a small reduction in catalytic efficiency and good stability. According to the FTIR results of the tested catalyst (Figure B), Mn6Ce0.3In0.7O showed a better water and sulfur resistance because indium distributed on the catalyst surface reduced the content of sulfate on the surface and protected the acidic site of MnCe. XRD test results (Figure ) showed that indium doping inhibited the crystallization of CeO2 and promoted the growth of the MnO2(100) crystal plane. The results showed that amorphous CeO2 and more MnO2(100) crystal surfaces were helpful to improve the activity of the catalyst. In the XRD results of different calcination temperatures, it can be found that the crystallization of the catalyst calcined at 300 °C was bad. The crystallization of the 400 °C calcined catalyst was excessive. The results showed that the crystallinity of the catalyst had a significant effect on the catalytic activity. It can be seen from the SEM images and specific surface area and pore size test results that the doping of indium changes the morphology and size of catalyst particles, increases the specific surface area and pore capacity of the catalyst, and thus provides more surface acid sites. XPS characterization was conducted to understand the surface element valence, surface oxygen state, acidity, and reducibility of the catalyst and then speculate the catalytic process of the catalytic reaction. The XPS test results show that the Mn6Ce0.3In0.7O catalyst has a higher proportion of Mn4+ and surface oxygen (Oα) than the Mn6CeO catalyst, which improves the catalytic oxide behavior and improves the SCR performance. In the Mn6Ce0.3In0.7O catalyst, the In3+ ions replace the position of Ce4+, resulting in the relative increase in the Ce3+ content and surface oxygen defects.[45] This oxygen vacancy can make the oxygen in the flue gas adsorbed and dissociated over the catalyst surface to produce oxygen with high flow and promote the oxidation of NO to NO2. In the reduction test, the reduction peak temperature of the Mn6Ce0.3In0.7O catalyst was reduced, indicating that the SCR activity temperature of the Mn6Ce0.3In0.7O catalyst was reduced, which is consistent with the calculation of the apparent activation energy. At around 700 °C, the hydrogen reduction curve of the Mn6Ce0.3In0.7O catalyst did not show the reduction peak of cerium, indicating that the oxidation of the catalyst was reduced to some extent and the N2O production was inhibited. In the test of surface acidity, the Mn6Ce0.3In0.7O catalyst showed enhanced surface acid site intensity and increased adsorption capacity for NH3, indicating that more NH3 was activated or decomposed on its surface and participated in the SCR denitrification reaction. In summary, the doping of the indium element increases the proportion of active centers due to the increase in the specific surface area of the catalyst. In addition, indium improves the intensity and the number of acidic sites to some extent and enhances the synergistic effect between multivalent cationic active centers in the SCR reaction, thus improving the catalytic efficiency.

Figure 10

Schematic flowchart of the catalytic activity in the evaluation device.

According to the existing reports,[27,61−63] the current mainstream view is that the low-temperature NH3-SCR reaction on the surface of a manganese-based catalyst follows the reaction process dominated by the Eley–Rideal (E–R) mechanism and supplemented by the Langmuir–Hinshelwood (L–H) mechanism. Therefore, the reaction process on the surface of the Mn6Ce0.3In0.7O catalyst was inferred according to XPS elemental valence analysis. It can be expressed as follows: NH3 is adsorbed forming Lewis acid sites to adsorb ammonia. NH3 is further dissociated to NH4+ and adsorbed NH2 on the surface of the catalyst forming Brønsted acid sites and reacts with gaseous NO or NO2 to generate the intermediate products NH2NO, NH4NO2, and NH4NO3. On the other hand, oxygen was activated combined with NO or NO2 in the raw gas at the oxygen defect at the catalyst surface and then reacted with adsorbed NH4+ at Brønsted acid sites to produce intermediate products NH4NO2 and NH4NO3. NH2NO and NH4NO2 were further decomposed into nitrogen and water due to instability, and NH4NO3 will further combine with NO to generate NH4NO2 and NO2 at low temperatures. As the temperature increases, NH4NO3 will decompose into H2O and harmful gas N2O. The possible reaction process on the surface of the Mn6Ce0.3In0.7O catalyst is given by the following:

Conclusions

MnCeInO catalysts were prepared for NH3-SCR by a coprecipitation method. The test results showed that Mn6Ce0.3In0.7O exhibited the best catalytic activity at low temperatures, and its optimal preparation temperature was 400 °C. The temperature window for denitrification efficiency greater than 90% (T90) was extended from 150–175 to 125–225 °C. The Mn6Ce0.3In0.7O catalyst had a low apparent activation energy (Ea) and a high turnover frequency (TOF) compared to the undoped catalyst without indium, indicating that the doping of indium improved the synergistic effect of the catalyst active sites. In addition, indium doping reduced the formation of sulfate on the catalyst and enhanced the water and sulfur resistance of the Mn6CeO catalyst. The denitrification efficiency of the Mn6Ce0.3In0.7O catalyst was higher than that of Mn6CeO by 15.4% after a 5 h test in the copresence of 5% H2O + 100 ppm SO2 in the feed gas. The enhanced redox performance and NH3 adsorption capacity of the Mn6Ce0.3In0.7O catalyst are related to the increase in the specific surface area of the catalyst and the increase in the ratio of Mn4+ ions and surface oxygen (Oα) on the surface.

Experimental Section

Catalysis Preparation

All samples were synthesized by coprecipitation. First, the MnCeO catalyst was synthesized, in which Mn:Ce = 6:1 (molar ratio). Next, a series of MnCeInO catalysts were prepared, in which Mn:(Ce + In) = 6:1 and Ce:In = 9:1, 7:3, 5:5, 3:7, and 1:9 (molar ratio). First, Mn(CH3COO)2·4H2O, In(NO3)3·H2O, and Ce(NO3)3·H2O were added one by one to 100 mL of distilled water at a water bath at 30 °C and stirred until a clear solution. An equal stoichiometric ratio of (NH3)2CO3 was dispersed in 10 mL of deionized water. It was then added into the clear salt solution. After continuous vigorous stirring, the mixture was adjusted to a pH value between 9 and 10 using 4 M NaOH solution and then covered with a cling film to be sealed, and the mixture was continued to be stirred for 2 h. Then, the suspension was allowed to stand at around 24 h at room temperature. The sediment was washed to pH = 7 using deionized water; then, the solid obtained was dried in a blast drying oven at 110 °C for 6 h. Finally, it was placed in a resistance furnace under an air atmosphere at 400 °C for 4 h. The prepared catalysts were named as Mn6Ce(In(O (a + b = 1). Mn6CeO was prepared by a similar approach without the inclusion of indium.

Catalyst Characterization and Computational Details

X-ray diffractometry (XRD) analysis for the samples was performed using a SmartLab (9 kW) rotating-target X-ray diffractometer of Keshikoshi Corporation (Japan).

In the Brunauer–Emmett–Teller (BET) test, the samples were pretreated by degassing at 180 °C and 5 mTorr under vacuum at a steady state. They were then tested at −196 °C using a Quantachrome Instruments Quadrasorb EVO. The catalyst morphology was observed by field emission scanning electron microscopy (FE-SEM) on a 15 kV Merlin compact device made by Carl Zeiss NTS GmbH (Germany).

The X-ray photoelectron spectra (XPS) of the samples were documented on a Thermo Fisher ESCALAB 250Xi spectrometer using monochromatic Al Kα as the X-ray source, calibrated with the C 1s peak of indeterminate carbon (binding energy of 284.8 eV). The splitting calculations of the Mn 2p peak, O 1s peak, Ce 3d peak, and In 3d peak were performed using XPSPEAK 4.1 splitting software for inverse folded products, with Shirley as the background and the convolution Gaussian/Lorentzian ratio set to 80/20.

Temperature-programmed desorption (NH3-TPD) of adsorbed NH3 on the catalyst was carried out on an AUTO Chem II 2920 device (American Microelectronics Instruments, Inc.) equipped with a thermal conductivity detector (TCD). A 0.12 g sample was loaded into a quartz TPD reactor and pretreated in a 50 mL/min stream of N2 and at 400 °C for 1 h. After waiting for the sample to cool to 50 °C, it was then purged with 10% NH3/N2 at a flow rate of 50 mL/min for 1 h to ensure complete saturation of the adsorption sites. The catalyst was then flushed with N2 at the same gas flux for 1 h to take out the weakly sorbed NH3. Then, the catalyst was heated from 50 to 800 °C with a constant N2 flow rate with a ramp rate of 10 °C/min for NH3 desorption, and the change in the NH3 content during the process was detected in real time using a TCD detector.

Hydrogen temperature-programmed reduction (H2-TPR) studies of mixed oxides were performed on an AUTO Chem II 2920 apparatus (American Microelectronics Instruments, Inc.) to determine their redox behavior. To perform these studies, a 0.12 g sample of the catalyst was put in a quartz U-shaped reaction cell. After the same sample pretreatment in pure argon as for NH3-TPD studies, they were cooled to 50 °C in a stream of argon. The TCD signals were recorded in the temperature range of 50–800 °C with a 5% H2/Ar mixture gas stream at a flow rate of 30 mL/min and a heating rate of 10 °C/min.

The surface adsorption of the catalyst was investigated by FTIR (Thermo Fisher) after the NH3-SCR reaction and water and sulfur resistance tests.

Catalytic Performance Test

The schematic diagram of experimental equipment is shown in Figure . The catalytic activity of Mn6CeO and Mn6Ce(In(O (a + b = 1) for NH3-SCR in excess oxygen was studied in a vertical tubular furnace with a high-temperature-resistant quartz glass tube of 6 mm inner diameter, and the catalyst was placed in the heating section of the tubular furnace. Catalysts (40–60 mesh) (0.2 g) with a volume of about 0.25 mL were used. The reaction gas consists of 500 ppm NO, 500 ppm NH3, 5% O2, 100 ppm SO2 (if used), 5% H2O (if used), and balance N2. The space velocity (GHSV) was about 120,000 h–1. The concentration of NO was measured by an electrochemical gas analyzer (Cairn-May Quintox Flue Gas Analyzer). The NO conversion and N2 selectivity were calculated using eqs and 2, respectively:[20]

The corner mark in represents the concentration of a substance in the raw gas. The corner mark out represents the concentration of a substance after catalyst treatment.

Turnover frequency (TOF) values were calculated according to the following equation:[21]where ν is the flow rate of nitrogen oxide (m3·s–1); α is the conversion of nitrogen oxide (%); Vm is the gas molar volume (m3·mol–1); na is the number of moles of surface acidic sites (mol). The TOF values based on the surface acidic sites were estimated by NH3-TPD.

The SCR kinetic parameters were calculated by the following equation:where k is the reaction rate constant (cm3·g–1·s–1), F is the total flow rate (cm3·s–1), W is the mass of the catalyst (g), and x is the NO conversion.

Furthermore, the apparent activation energies (Ea) was calculated using the Arrhenius equation shown as follows: