Enhanced Low Temperature NO Reduction Performance via MnO xFe 2 O 3 / Vermiculite Monolithic Honeycomb Catalysts

Selective catalytic reduction of NOx by ammonia (NH3-SCR) was the most efficient and economic technology for De-NOx applications. Therefore, a series of MnOx/vermiculite (VMT) and MnOx-Fe2O3/VMT catalysts were prepared by an impregnation method for the selective catalytic reduction (SCR) of nitrogen oxides (NOx). The MnOx-Fe2O3/VMT catalysts provided an excellent NO conversion of 96.5% at 200 ◦C with a gas hourly space velocity (GHSV) of 30,000 h−1 and an NO concentration of 500 ppm. X-ray photoelectron spectroscopy results indicated that the Mn and Fe oxides of the MnOx-Fe2O3/VMT catalyst were mainly composed of MnO2 and Fe2O3. However, the MnO2 and Fe2O3 components were well dispersed because no discernible MnO2 and Fe2O3 phases were observed in X-ray powder diffraction spectra. Corresponding MnOx-Fe2O3/VMT monolithic honeycomb catalysts (MHCs) were prepared by an extrusion method, and the MHCs achieved excellent SCR activity at low temperature, with an NO conversion greater than 98.6% at 150 ◦C and a GHSV of 4000 h−1. In particular, the MnOx-Fe2O3/VMT MHCs provided a good SCR activity at room temperature (20 ◦C), with an NO conversion of 62.2% (GHSV = 1000 h−1). In addition, the NO reduction performance of the MnOx-Fe2O3/VMT MHCs also demonstrated an excellent SO2 resistance.


Introduction
Nitrogen oxides (NO x ) in stationary stack source emissions are strong contributing factors of acid rain, photochemical smog, and ozone depletion and are, therefore, detrimental to the natural environment and human health [1][2][3].Therefore, the selective catalytic reduction (SCR) of NO x by ammonia (NH 3 -SCR) was developed to reduce the release of NO x (i.e., De-NO x ), and this has thus far been the most economical and effective technology for De-NO x applications [4].It is well known that the key component affecting the SCR and economic performance of NH 3 -SCR technology is the catalyst employed in the process, which should be low in cost, and provide high SCR activity at low temperatures.As a result, manganese-based catalysts have been widely studied over the past few decades owing to their low cost and promisingly high SCR activity at low temperature [5].However, the SCR activity of pure MnO x catalysts is profoundly deteriorated by SO 2 and H 2 O, and emissions with high concentrations of SO 2 and H 2 O can lead to poisoning and the eventual deactivation of pure MnO x catalysts.These drawbacks can be overcome, and the SCR activity of pure MnO x catalysts can even be improved by incorporating transition metal or rare earth elements as catalyst promoters [6,7].For example, Chen et al. reported that Fe-MnO x mixed-oxide catalysts provided a 100% NO x conversion in the temperature range from 140 • C to 220 • C [8].
Since SCR activity occurs on the surfaces of catalysts, the SCR activity of catalysts can be greatly enhanced, particularly under low-temperature conditions, by maximizing their surface area via distribution over high surface area supports.Numerous types of supports have been employed for Mn-based catalysts, such as molecular sieves, carbon materials, and metal oxides.Molecular sieves modified with Mn-based catalyst materials have shown excellent SCR activity because of their regular pore structures, high strength, and high specific surface areas [9].Numerous Mn-based molecular sieve supported catalysts, such as Mn supported on ZSM-5 zeolite (denoted as Mn/ZSM-5) [10] and MnO x /silicoaluminophosphate zeolite (SAPO-34) [11], have been shown to be highly active for NO x reduction at low temperatures.Carbon materials have also been widely used as catalyst supports because of their well-developed porosity, high specific surface area, chemical stability, strong adsorption ability, and excellent thermal conductivity [12].For example, Lu et al. synthesized MnO 2 supported on carbon nanotubes (CNTs), and the catalyst attained an NO conversion of 89.5% at 180 • C [13].Finally, metal oxide supports provide high catalyst surface areas, high thermal stability, and surface acid-base properties [14,15].Manganese-based catalysts supported on metal oxides, such as MnO x /TiO 2 [16][17][18], MnO x /Al 2 O 3 [19], MnO x /CeO 2 [20][21][22], and MnO x /Ce-ZrO 2 [23], have generated considerable attention, and represent promising catalysts for the SCR of NO x at low temperatures.
In recent years, many researchers have focused on catalysts employing supports composed of natural ore materials, such as montmorillonite, saponite, attapulgite, chabazite, and vermiculite (VMT), due to their great abundance, which makes them easily obtainable and inexpensive materials.A number of catalysts have incorporated these natural ore materials, such as Zr-Mn/attapulgite (ATP) [24], Ce-MnO x /ATP [25], porous clay hetero-structures (PCH) modified with NH 3 -Cu [26], Cu-chabazite (CHA) [27][28][29] and crystalline MnO 2 (c-MnO 2 )/TiO 2 -palygorskite (PAL) [30].Among the available natural ore materials, the unique layered structure of VMT makes it an excellent candidate as a support for the SCR of NO.In addition, VMT includes metal oxides that can actively assist in catalytic reactions, and its great abundance in Xinjiang, China makes this material of particular interest to Chinese researchers.[32].Moreover, VMT has been widely investigated for use in heterogeneous catalysts, such as in the photocatalyst TiO 2 /VMT [33], NiO/VMT for carbon monoxide methanation [34,35], and a novel HgCl 2 /EML-VMT-C catalyst employing expanded multilayered (EML) VMT mixed with carbon on the surface (EML-VMT-C) for the hydrochlorination of acetylene [36].However, the application of catalysts employing VMT supports for the SCR of NO x at low temperatures remains challenging.
In light of the substantial cost benefits associated with the use of natural ore materials as catalyst supports, the present study employs VMT as supports in the fabrication of MnO x -Fe 2 O 3 /VMT catalysts by the impregnation method for the SCR of NO by NH 3 .The as-prepared MnO x -Fe 2 O 3 /VMT catalysts provide an excellent NO conversion of 96.5% at 200 • C with a gas hourly space velocity (GHSV) of 30,000 h −1 .To improve the low temperature performance, corresponding MnO x -Fe 2 O 3 /VMT monolithic honeycomb catalysts (MHCs) were prepared by an extrusion method, and the MHCs achieve an NO conversion greater than 98.6% at 150 • C and GHSV = 4000 h −1 .In particular, the MnO x -Fe 2 O 3 /VMT MHCs provided good SCR activity at room temperature (20 • C), with an NO conversion of 62.2% (GHSV = 1000 h −1 ).Finally, because flue gases still contain small amounts of SO 2 after desulfurization and water removal, we also investigated the influence of SO 2 on the catalytic performance of MnO x -Fe 2 O 3 /VMT MHCs.

Results and Discussion
The results of catalytic activity testing for the as-prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts are shown in Figure 1.As shown in Figure 1a, the NO conversion of the MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts increased obviously with increasing temperature in the low temperature region until attaining a maximum value, after which the NO conversion decreased.The highest NO conversion attained for the MnO x -Fe 2 O 3 /VMT catalyst was 96.5% at 200 • C, while the highest NO conversion attained for the MnO x /VMT catalyst was 93.1% at 250 • C. Compared with the MnO x /VMT catalyst, the catalytic activity of the MnO x -Fe 2 O 3 /VMT catalyst was between 6% and 16% greater than that of the MnO x /VMT catalyst in the low temperature range of 20-200 • C, indicating that the addition of Fe obviously increased the low temperature activity of the catalyst.For both catalysts, the declining NO conversion at relatively high temperatures originated from the oxidation of the ammonia.
Catalysts 2018, 8, x FOR PEER REVIEW 3 of 14 after desulfurization and water removal, we also investigated the influence of SO2 on the catalytic performance of MnOx-Fe2O3/VMT MHCs.

Results and Discussion
The results of catalytic activity testing for the as-prepared MnOx/VMT and MnOx-Fe2O3/VMT catalysts are shown in Figure 1.As shown in Figure 1a, the NO conversion of the MnOx/VMT and MnOx-Fe2O3/VMT catalysts increased obviously with increasing temperature in the low temperature region until attaining a maximum value, after which the NO conversion decreased.The highest NO conversion attained for the MnOx-Fe2O3/VMT catalyst was 96.5% at 200 °C, while the highest NO conversion attained for the MnOx/VMT catalyst was 93.1% at 250 °C.Compared with the MnOx/VMT catalyst, the catalytic activity of the MnOx-Fe2O3/VMT catalyst was between 6% and 16% greater than that of the MnOx/VMT catalyst in the low temperature range of 20-200 °C, indicating that the addition of Fe obviously increased the low temperature activity of the catalyst.For both catalysts, the declining NO conversion at relatively high temperatures originated from the oxidation of the ammonia.In addition, Figure 1b shows that the N2 selectivity of the MnOx/VMT and MnOx-Fe2O3/VMT catalysts gradually declined with increasing temperature in the low temperature region (20-100 °C), and sharply declined in the middle temperature region (150-200 °C).The MnOx-Fe2O3/VMT catalyst exhibited excellent N2 selectivities of 97.1% and 95.9% at 20 °C and 50 °C, respectively, and its N2 selectivity was greater than that of the MnOx/VMT catalyst at temperatures less than 250 °C.However, we note that the sharp drop in the N2 selectivity of the MnOx-Fe2O3/VMT catalyst at 150 °C resulted in an N2 selectivity that was less than that of the MnOx/VMT catalyst at temperatures greater than 250 °C.The N2 selectivity results are readily correlated with the measured N2O and NO2 contents shown in Figure 1c,d, respectively.Here, we note that the gradual decline in the N2 selectivity of both catalysts at low temperature corresponds with a gradually increasing generation of N2O and NO2 In addition, Figure 1b shows that the N 2 selectivity of the MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts gradually declined with increasing temperature in the low temperature region (20-100 • C), and sharply declined in the middle temperature region (150-200 • C).The MnO x -Fe 2 O 3 /VMT catalyst exhibited excellent N 2 selectivities of 97.1% and 95.9% at 20 • C and 50 • C, respectively, and its N 2 selectivity was greater than that of the MnO x /VMT catalyst at temperatures less than 250 • C.However, we note that the sharp drop in the N 2 selectivity of the MnO x -Fe 2 O 3 /VMT catalyst at 150 • C resulted in an N 2 selectivity that was less than that of the MnO x /VMT catalyst at temperatures greater than 250 • C. The N 2 selectivity results are readily correlated with the measured N 2 O and NO 2 contents shown in Figure 1c,d, respectively.Here, we note that the gradual decline in the N 2 selectivity of both catalysts at low temperature corresponds with a gradually increasing generation of N 2 O and NO 2 over a similar temperature range.In addition, the sharply declining N 2 selectivity of both catalysts in the middle temperature region is mainly because the N 2 O content is increasing rapidly for temperatures greater than 150 • C. We also note that the region over which the N 2 selectivity of the MnO x /VMT catalyst was greater than that of the MnO x -Fe 2 O 3 /VMT catalyst (i.e., at temperatures greater than 250 • C) corresponds with the fact that both the N 2 O and the NO 2 contents were less for the MnO x /VMT catalyst than for the MnO x -Fe 2 O 3 /VMT catalyst at temperatures greater than 250 • C. Nevertheless, we note that the MnO x -Fe 2 O 3 /VMT catalyst demonstrated both a better NO conversion and a better N 2 selectivity that those of the MnO x /VMT catalyst in the temperature range of 20-200 • C.
Figure 2a presents X-ray diffractometer (XRD) patterns of the VMT support and as-prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT powdered catalyst samples.We note that VMT presents several strong peaks that exhibit greatly decreased intensities after impregnation.In addition, several diffraction peaks indicative of Mn 3 O 4 (PDF#18-0803) are observed for the MnO x /VMT catalyst at 18.1 • , 29.2 • , 32.4 • , 36.2 • , 51.0 • , and 58.5 • , while no others crystal phases of MnO x are evident.These results suggest that the existence of diffraction peaks could be due to large crystals of MnO x , resulting in weak XRD peaks indicative of MnO x and VMT [37].The MnO x -Fe 2 O 3 /VMT catalyst sample presents even weaker XRD peaks indicative of MnO x and VMT, and no additional crystal phases are observed.
Here, the coexistence of manganese and iron oxides enhances dispersion, and consequently reduces the crystallinity, indicating the presence of strong interactions between these two metal oxides [38].
Catalysts 2018, 8, x FOR PEER REVIEW 4 of 14 over a similar temperature range.In addition, the sharply declining N2 selectivity of both catalysts in the middle temperature region is mainly because the N2O content is increasing rapidly for temperatures greater than 150 °C.We also note that the region over which the N2 selectivity of the MnOx/VMT catalyst was greater than that of the MnOx-Fe2O3/VMT catalyst (i.e., at temperatures greater than 250 °C) corresponds with the fact that both the N2O and the NO2 contents were less for the MnOx/VMT catalyst than for the MnOx-Fe2O3/VMT catalyst at temperatures greater than 250 °C.Nevertheless, we note that the MnOx-Fe2O3/VMT catalyst demonstrated both a better NO conversion and a better N2 selectivity that those of the MnOx/VMT catalyst in the temperature range of 20-200 °C. Figure 2a presents X-ray diffractometer (XRD) patterns of the VMT support and as-prepared MnOx/VMT and MnOx-Fe2O3/VMT powdered catalyst samples.We note that VMT presents several strong peaks that exhibit greatly decreased intensities after impregnation.In addition, several diffraction peaks indicative of Mn3O4 (PDF#18-0803) are observed for the MnOx/VMT catalyst at 18.1°, 29.2°, 32.4°, 36.2°,51.0°, and 58.5°, while no others crystal phases of MnOx are evident.These results suggest that the existence of diffraction peaks could be due to large crystals of MnOx, resulting in weak XRD peaks indicative of MnOx and VMT [37].The MnOx-Fe2O3/VMT catalyst sample presents even weaker XRD peaks indicative of MnOx and VMT, and no additional crystal phases are observed.
Here, the coexistence of manganese and iron oxides enhances dispersion, and consequently reduces the crystallinity, indicating the presence of strong interactions between these two metal oxides [38].

Figure 2b-d present SEM micrographs indicative of the morphologies of the VMT support and as-prepared MnO
x /VMT and MnO x -Fe 2 O 3 /VMT catalysts, respectively.We note the distinct layered structure of the VMT support, and that the layer surfaces are very smooth, without obvious pores or wrinkles.After impregnation, the MnO x /VMT catalyst exhibits a distribution of irregular loose particles, and the layer surfaces of the VMT support appear to be very rough (Figure 2c), indicating the formation of many new channels on the VMT surfaces.However, the sizes of the irregular loose particles of the MnO x -Fe 2 O 3 /VMT catalyst are substantially decreased.These results explain the reason for the increased surface area and decreased pore diameter of the MnO x -Fe 2 O 3 /VMT catalyst relative to those of the MnO x /VMT catalyst.
The N 2 adsorption-desorption isotherm plots and the corresponding BJH pore size distribution curves of the VMT support and as-prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalyst samples are presented in Figure 2e,f, respectively.From Figure 2e, we note the presence of well-defined type II hysteresis loops with sloping adsorption branches for all samples, which is particularly pronounced for the MnO x -Fe 2 O 3 /VMT catalyst.From Figure 2f, we note that both catalyst samples exhibit three narrow peaks in the pore diameter range of 2-10 nm.The textural data for all samples are listed in Table 1.From the table, we find that the impregnation of 20 wt% Mn more than doubled the BET surface area of the MnO x /VMT catalyst sample relative to that of the VMT support, but, surprisingly, the pore diameter decreased by about 30%.Moreover, the impregnation of 20 wt% Mn and 5 wt% Fe further increased the BET surface area of the MnO x -Fe 2 O 3 /VMT catalyst sample by a factor greater than 3 relative to the MnO x /VMT catalyst sample, while the pore diameter further decreased by about 12%.These results indicate that the doping of second metal can change the surface structure of the MnO x -Fe 2 O 3 /VMT catalyst [39], and the calcination subsequent to impregnation may have formed additional channels in the surfaces of the VMT support.A high BET surface area is beneficial toward increasing the number of active sites of a catalyst and, thus, provides an increased NO conversion.The XPS spectra of the as-prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts are shown in Figure 3, and their principle surface compositions obtained from the fitted spectra in Figure 3b-d are listed in Table 2. Sharp photoelectron peaks are observed in Figure 3a for Fe, Mn, O, and C elements with binding energies of 712.1 eV (Fe 2p 3/2 ), 642.1 eV (Mn 2p 3/2 ), 532.1 eV (O 1s), and 284.1 eV (C 1s), respectively.We note that the XPS spectrum obtained for the MnO x /VMT catalyst also exhibits a small peak indicative of Fe because the VMT support material naturally contains a small concentration of Fe.The Fe 2p, Mn 2p, and O 1s spectra of the catalyst samples are individually discussed in detail below.about 37% from the MnOx/VMT catalyst to the MnOx-Fe2O3/VMT catalyst.This indicates that the addition of Fe facilitates the conversion of MnOx to MnO2 on the catalyst surface.It has been reported [41][42][43] that the NO conversion capability of pure manganese oxides can be ranked as MnO2 > Mn5O8 > Mn2O3 > Mn3O4.In addition, it has been reported that a greater concentration of MnO2 on the catalyst surface promotes the SCR reaction [44].Therefore, it can be expected that the MnOx-Fe2O3/VMT catalyst will provide an improved NO conversion relative to that of the MnOx/VMT catalyst.The Fe 2p spectra presented in Figure 3c exhibit electron binding energy peak values for Fe 2p3/2 and Fe 2p1/2 states, and a satellite peak of 710.7, 725.0, and 718.6 eV, respectively.The satellite peak energy corresponds well with that reported for Fe2O3 [45].The Fe 2p3/2 peak was deconvolved into two components with peaks at 710.4 eV and 712.6 eV indicative of Fe 2+ and Fe 3+ phases [46], respectively.The results indicate that the percentage of Fe atoms in the Fe 3+ state listed in Table 2 are about 13% less in the VMT support of the MnOx/VMT catalyst than that of the MnOx-Fe2O3/VMT catalyst.According to a past study [47], Fe 3+ sites may facilitate the reduction of NOx at low temperature.Thus, the MnOx-Fe2O3/VMT catalyst can be expected to provide a slightly better lowtemperature activity than that of the MnOx/VMT catalyst.
As shown in Figure 3d the O 1s peaks of the MnOx/VMT and MnOx-Fe2O3/VMT catalysts at 528-535 eV were deconvolved into three peaks, denoted as Olatt (529.9 eV), Oads (531.4 eV), and Osur As shown in Figure 3b, the Mn 2p spectra include two main peaks at binding energies of 653.8 ± 0.4 eV and 641.9 ± 0.4 eV, which are assigned to Mn 2p 3/2 and Mn 2p 1/2 electron states, respectively.To identify the specific Mn species of each sample, the Mn 2p 3/2 peak was deconvoluted into three peaks, corresponding to Mn 2+ (641.0 ± 0.4 eV), Mn 3+ (642.1 ± 0.4 eV), and Mn 4+ (643.5 ± 0.4 eV), respectively [40].The percentage of Mn atoms in the Mn 4+ state listed in Table 2 was then determined as the area under the curve representative of Mn 4+ relative to the total area under the Mn 2p 3/2 curve.These values are indicative of the molar concentration of MnO 2 relative to all MnO x on the surfaces of the MnO x -Fe 2 O 3 /VMT and MnO x /VMT catalysts.We note that the Mn 4+ percentage increased by about 37% from the MnO x /VMT catalyst to the MnO x -Fe 2 O 3 /VMT catalyst.This indicates that the addition of Fe facilitates the conversion of MnO x to MnO 2 on the catalyst surface.It has been reported [41][42][43] that the NO conversion capability of pure manganese oxides can be ranked as MnO 2 > Mn 5 O 8 > Mn 2 O 3 > Mn 3 O 4 .In addition, it has been reported that a greater concentration of MnO 2 on the catalyst surface promotes the SCR reaction [44].Therefore, it can be expected that the MnO x -Fe 2 O 3 /VMT catalyst will provide an improved NO conversion relative to that of the MnO x /VMT catalyst.
The Fe 2p spectra presented in Figure 3c exhibit electron binding energy peak values for Fe 2p 3/2 and Fe 2p 1/2 states, and a satellite peak of 710.7, 725.0, and 718.6 eV, respectively.The satellite peak energy corresponds well with that reported for Fe 2 O 3 [45].The Fe 2p 3/2 peak was deconvolved into two components with peaks at 710.4 eV and 712.6 eV indicative of Fe 2+ and Fe 3+ phases [46], respectively.The results indicate that the percentage of Fe atoms in the Fe 3+ state listed in Table 2 are about 13% less in the VMT support of the MnO x /VMT catalyst than that of the MnO x -Fe 2 O 3 /VMT catalyst.
According to a past study [47], Fe 3+ sites may facilitate the reduction of NO x at low temperature.Thus, the MnO x -Fe 2 O 3 /VMT catalyst can be expected to provide a slightly better low-temperature activity than that of the MnO x /VMT catalyst.
As shown in Figure 3d the O 1s peaks of the MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts at 528-535 eV were deconvolved into three peaks, denoted as O latt (529.9 eV), O ads (531.4 eV), and O sur (532.4 eV), which are attributed to O atoms bonded with metal cations, in adsorbed water, and in surface hydroxyl groups, respectively [48].It has been widely reported that oxygen in the gas phase can be activated by oxygen vacancies on the surface of SCR catalysts.Therefore, the relative abundance of O ads is of particular interest because an increased percentage of O ads can promote the oxidation of NO to NO 2 , and enhance the SCR performance at low temperature through a rapid NH 3 -SCR route [49,50].As shown in Table 2, the ratio of O ads /O total on the surface of the MnO x -Fe 2 O 3 /VMT catalyst is about 19% greater than that for the MnO x /VMT catalyst.As a result, we can expect this factor to further enhance the low-temperature activity of the MnO x -Fe 2 O 3 /VMT catalyst relative to that of the MnO x /VMT catalyst.
The results of H 2 -TPR testing are presented in Figure 4 for the VMT support and the as-prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalysts.We note that several peaks are observable for all samples within the temperature range of 200-800 • C. The H 2 -TPR curve for the VMT support includes only two peaks, with a reduction peak attributed to Fe 2 O 3 → Fe 3 O 4 at 274 • C, and a second peak located at 688 • C that may be attributed to Fe 3 O 4 → FeO [51].Compared with the VMT support curve, two new reduction peaks are observed in the H 2 -TPR curve for the MnO x /VMT catalyst due to the addition of Mn.From previous studies [52], the reduction peaks of MnO x can be assigned to the reduction processes of MnO 2 via Mn 2 O 3 to MnO.Here, the first peak observed at 269 • C can be attributed to the MnO 2 → Mn 2 O 3 reduction transition, while the peak at 469 • C can be attributed to Mn 2 O 3 → MnO [53].Therefore, because the MnO x /VMT catalyst includes Fe 2 O 3 in the support, the peak centered at 269 • C can be attributed to both Fe 2 O 3 → Fe 3 O 4 and MnO 2 → Mn 2 O 3 .Meanwhile, the third peak for the MnO x /VMT sample located at 608 • C can be attributed to Fe 3 O 4 → FeO, and the forth peak located at 742 • C can be attributed to FeO → Fe [54].The H 2 -TPR curve for the MnO x -Fe 2 O 3 /VMT catalyst exhibited similar reduction peaks above 260 • C, but which were shifted to lower temperatures relative to those for the MnO x /VMT catalyst, suggesting that the Fe and Mn species in the MnO x -Fe 2 O 3 /VMT catalyst were more easily reduced.This can be ascribed to the previously discussed synergetic effect between Mn and Fe, which could effectively promote the redox properties of the MnO x -Fe 2 O 3 /VMT catalyst and improve its catalytic activity.
Catalysts 2018, 8, x FOR PEER REVIEW 7 of 14 (532.4eV), which are attributed to O atoms bonded with metal cations, in adsorbed water, and in surface hydroxyl groups, respectively [48].It has been widely reported that oxygen in the gas phase can be activated by oxygen vacancies on the surface of SCR catalysts.Therefore, the relative abundance of Oads is of particular interest because an increased percentage of Oads can promote the oxidation of NO to NO2, and enhance the SCR performance at low temperature through a rapid NH3-SCR route [49,50].As shown in Table 2, the ratio of Oads/Ototal on the surface of the MnOx-Fe2O3/VMT catalyst is about 19% greater than that for the MnOx/VMT catalyst.As a result, we can expect this factor to further enhance the low-temperature activity of the MnOx-Fe2O3/VMT catalyst relative to that of the MnOx/VMT catalyst.
The results of H2-TPR testing are presented in Figure 4 for the VMT support and the as-prepared MnOx/VMT and MnOx-Fe2O3/VMT catalysts.We note that several peaks are observable for all samples within the temperature range of 200-800 °C.The H2-TPR curve for the VMT support includes only two peaks, with a reduction peak attributed to Fe2O3 → Fe3O4 at 274 °C, and a second peak located at 688 °C that may be attributed to Fe3O4 → FeO [51].Compared with the VMT support curve, two new reduction peaks are observed in the H2-TPR curve for the MnOx/VMT catalyst due to the addition of Mn.From previous studies [52], the reduction peaks of MnOx can be assigned to the reduction processes of MnO2 via Mn2O3 to MnO.Here, the first peak observed at 269 °C can be attributed to the MnO2 → Mn2O3 reduction transition, while the peak at 469 °C can be attributed to Mn2O3 → MnO [53].Therefore, because the MnOx/VMT catalyst includes Fe2O3 in the support, the peak centered at 269 °C can be attributed to both Fe2O3 → Fe3O4 and MnO2 → Mn2O3.Meanwhile, the third peak for the MnOx/VMT sample located at 608 °C can be attributed to Fe3O4 → FeO, and the forth peak located at 742 °C can be attributed to FeO → Fe [54].The H2-TPR curve for the MnOx-Fe2O3/VMT catalyst exhibited similar reduction peaks above 260 °C, but which were shifted to lower temperatures relative to those for the MnOx/VMT catalyst, suggesting that the Fe and Mn species in the MnOx-Fe2O3/VMT catalyst were more easily reduced.This can be ascribed to the previously discussed synergetic effect between Mn and Fe, which could effectively promote the redox properties of the MnOx-Fe2O3/VMT catalyst and improve its catalytic activity.A photograph of the as-prepared MnOx-Fe2O3/VMT MHCs is shown in Figure 5a.The macrostructure of the MnOx-Fe2O3/VMT MHCs is important for promoting catalytic activity in industrial applications, and their primary physical properties have significant effects on their mechanical and catalytic performances.The SEM micrograph of an as-prepared MnOx-Fe2O3/VMT MHCs in Figure 5b shows that the glass fibers were uniformly distributed throughout the MnOx-Fe2O3/VMT MHCs, which can be expected to provide enhanced mechanical strength.The XRD A photograph of the as-prepared MnO x -Fe 2 O 3 /VMT MHCs is shown in Figure 5a.The macrostructure of the MnO x -Fe 2 O 3 /VMT MHCs is important for promoting catalytic activity in industrial applications, and their primary physical properties have significant effects on their mechanical and catalytic performances.The SEM micrograph of an as-prepared MnO x -Fe 2 O 3 /VMT MHCs in Figure 5b shows that the glass fibers were uniformly distributed throughout the MnO x -Fe 2 O 3 /VMT MHCs, which can be expected to provide enhanced mechanical strength.The XRD pattern of a representative MnO x -Fe 2 O 3 /VMT MHCs is shown in Figure 5c.In contrast with the XRD results for the MnO x -Fe 2 O 3 /VMT powdered catalyst (Figure 2a), the bentonite content in the MHCs, which has a large proportion of SiO 2 , yields a sharp peak attributable to SiO 2 (PDF#27-0605) at 21.6 • .In addition, two sharp peaks attributable to carbon (PDF#41-1487) are observed at 26.4 • and 54.5 • owing to the decomposition of the organic additives.Figure 5d presents the N 2 isotherms and corresponding pore size distribution curve (inset) of a MnO x -Fe 2 O 3 /VMT MHCs, and the corresponding textural data are listed Table 1.Here, we note that, while the BET surface area and poor volume of the MnO x -Fe 2 O 3 /VMT MHCs were less than those of its powdered counterpart, the average pore diameter was increased by around 10%.The porous texture of the MnO x -Fe 2 O 3 /VMT MHCs can be expected to play a significant role in the SCR of NO x owing to an enhanced transportation and adsorption of reactant gases.
Catalysts 2018, 8, x FOR PEER REVIEW 8 of 14 which has a large proportion of SiO2, yields a sharp peak attributable to SiO2 (PDF#27-0605) at 21.6°.
In addition, two sharp peaks attributable to carbon (PDF#41-1487) are observed at 26.4° and 54.5° owing to the decomposition of the organic additives.Figure 5d presents the N2 isotherms and corresponding pore size distribution curve (inset) of a MnOx-Fe2O3/VMT MHCs, and the corresponding textural data are listed Table 1.Here, we note that, while the BET surface area and poor volume of the MnOx-Fe2O3/VMT MHCs were less than those of its powdered counterpart, the average pore diameter was increased by around 10%.The porous texture of the MnOx-Fe2O3/VMT MHCs can be expected to play a significant role in the SCR of NOx owing to an enhanced transportation and adsorption of reactant gases.The NO conversion of MnOx-Fe2O3/VMT MHCs at different GHSV values is shown as a function of reaction temperature in Figure 6a.We note that the NO conversion decreased with an increasing GHSV from 4000 h −1 to 8000 h −1 .The temperature region of highest activity was between 150 °C and 200 °C, and the NO conversion attained a maximum value greater than 98% with GHSV = 4000 h −1 .Meanwhile, the MnOx-Fe2O3/VMT MHCs exhibited an excellent NO conversion of 39.2% and 53.4% at 20 °C and 50 °C, respectively, with GHSV = 4000 h −1 .The cycling stabilities of MnOx-Fe2O3/VMT MHCs are presented in Figure 6b at various temperatures with different GHSV values.We note that the NO conversion was stable, with no significant changes over a full 10 h of testing.We also find that the MnOx-Fe2O3/VMT MHCs provided excellent NO conversion values of 98.6% and 85.8% at 150 °C and 100 °C, respectively, with GHSV = 4000 h −1 .The results obtained are comparable to those reported in the literature (Figure 6c).Thus, a NO conversion of 92% at 500 °C has been reported for −1 The NO conversion of MnO x -Fe 2 O 3 /VMT MHCs at different GHSV values is shown as a function of reaction temperature in Figure 6a.We note that the NO conversion decreased with an increasing GHSV from 4000 h −1 to 8000 h −1 .The temperature region of highest activity was between 150 • C and 200 • C, and the NO conversion attained a maximum value greater than 98% with GHSV = 4000 h −1 .Meanwhile, the MnO x -Fe 2 O 3 /VMT MHCs exhibited an excellent NO conversion of 39.2% and 53.4% at 20 • C and 50 • C, respectively, with GHSV = 4000 h −1 .The cycling stabilities of MnO x -Fe 2 O 3 /VMT MHCs are presented in Figure 6b at various temperatures with different GHSV values.We note that the NO conversion was stable, with no significant changes over a full 10 h of testing.We also find that the MnO x -Fe 2 O 3 /VMT MHCs provided excellent NO conversion values of 98.6% and 85.8% at 150 • C and 100 • C, respectively, with GHSV = 4000 h −1 .The results obtained are comparable to those reported in the literature (Figure 6c).Thus, a NO conversion of 92% at 500 • C has been reported for WO 3 -TiO 2 (GHSV of 11,000 h −1 ) [55], a conversion of 98% at 400 • C was reached by V 2 O 5 -MoO 3 /TiO 2 (GHSV of 6000 h −1 ) [56].A TiO 2 catalyst showed a conversion of 80% at 300 • C (GHSV of 25,000 h −1 ) [57], 75% at 250 • C for a V 2 O 5 -WO 3 /TiO 2 catalyst (GHSV of 27,000 h −1 ) [58], and 90% at 160 • C for a Cr-V/TiO 2 catalyst (GHSV of 4000 h −1 ) [59].Even at GHSV = 2000 h −1 , the MnO  The impact of SO2 on the cycling performance of MnOx-Fe2O3/VMT MHCs at different temperatures and GHSV values is shown in Figure 6d.After the first 1 h of testing when 300 ppm SO2 was introduced into the reaction gas, we note that the extent to which the NO conversion decreased was only slight at 20 °C (GHSV = 1000 h −1 ), but increased with increasing temperature from 50 °C (GHSV = 2000 h −1 ) to 150 °C (GHSV = 4000 h −1 ).This is because the crystallization temperature of sulfate is greater than 40 °C.We also note that the NO conversion of the MnOx-Fe2O3/VMT MHCs was stable over the entire period of SO2 addition, which indicates good SO2 resistance.However, the NO conversion values obtained under all conditions did not recover to their original values when the addition of SO2 was discontinued.These results suggest that the decreased activity of the MHCs was not due to the competitive adsorption of SO2, but because of the formation of sulfates covering the active sites of the catalysts [60].The impact of SO 2 on the cycling performance of MnO x -Fe 2 O 3 /VMT MHCs at different temperatures and GHSV values is shown in Figure 6d.After the first 1 h of testing when 300 ppm SO 2 was introduced into the reaction gas, we note that the extent to which the NO conversion decreased was only slight at 20 • C (GHSV = 1000 h −1 ), but increased with increasing temperature from 50 • C (GHSV = 2000 h −1 ) to 150 • C (GHSV = 4000 h −1 ).This is because the crystallization temperature of sulfate is greater than 40 • C. We also note that the NO conversion of the MnO x -Fe 2 O 3 /VMT MHCs was stable over the entire period of SO 2 addition, which indicates good SO 2 resistance.However, the NO conversion values obtained under all conditions did not recover to their original values when the addition of SO 2 was discontinued.These results suggest that the decreased activity of the MHCs was not due to the competitive adsorption of SO 2 , but because of the formation of sulfates covering the active sites of the catalysts [60].Vermiculite supports were prepared by a microwave method as follows.The raw VMT (Xinjiang Yuli Xinlong Vermiculite Co., Ltd., Korla, China) was washed with water until no trace of foreign material was observable under visual inspection, and then dried in an oven at 100 • C. Finally, the washed VMT was placed in a 500 mL beaker and expanded in a microwave.The VMT was collected and placed in a sealed container, and then crushed prior to use.All the catalysts were prepared by the impregnation method.We completely dissolved Mn(CH 3 COO) 2 •4H 2 O in water under stirring in an appropriate ratio to the VMT content (i.e., 20 wt% Mn).We added the VMT powder, and continued stirring for 10 h.The sample was dried in air at 100 • C for 12 h, and then crushed and sieved in an 80-100 mesh sieve.Finally, the sample was calcined in air at 500 • C for 5 h.We prepared MnO x -Fe 2 O 3 /VMT catalysts by an equivalent method using Mn(CH 3 COO) 2 •4H 2 O and Fe(NO 3 ) 3 •9H 2 O (20 wt% Mn-5 wt% Fe relative to VMT).

Preparation of MnO x -Fe 2 O 3 /VMT Monolithic Honeycomb Catalysts
We synthesized MnO x -Fe 2 O 3 /VMT MHCs by an extrusion molding method.The powdered MnO x -Fe 2 O 3 /VMT catalyst was dry mixed with 10 wt% bentonite and 3 wt% carboxy methyl cellulose (CMC) in a blender mixer.The materials in the blender mixer were then wet mixed with 10 wt% glycerin, followed by a sufficient amount of water to ensure an appropriate viscosity for extrusion.The resulting mixture was kneaded by hand for 30 min until achieving a uniform consistency.The sample was subjected to vacuum de-airing and aged for 24 h to increase the plasticity.The aged sample was molded by an extruding machine to obtain a monolithic honeycomb structure.The honeycomb sample was dried at 70 • C in air for 12 h in a muffle furnace, further dried at 100 • C for 12 h, and then calcined at 500 • C for 5 h.The rate of temperature increase was 1 • C/min for all the above steps.The mold size was 3.3 cm × 3.3 cm with 36 channels (i.e., 6 × 6 cells).

Catalyst Characterization
The morphologies of the powder catalysts and MHCs were characterized by scanning electron microscopy (SEM; Hitachi S-4300, Hitachi Limited, Tokyo, Japan).A BET apparatus (Micromeritics ASAP 2020, Micromeritics Instrument Ltd., Norcross, GA, USA) was employed to measure the Brunauer-Emmett-Teller (BET) specific surface area and Barrett-Joyner-Halenda (BJH) pore structure of the powder catalysts and MHCs.The samples were degassed in vacuum at 200 • C for 4 h prior to measurement.The total pore volume was calculated from the volume of nitrogen adsorbed at P/P 0 = 0.99.Powder X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance X-ray diffractometer (Bruker Biosciences Corporation, Billerica, MA, USA) with Cu Kα radiation (λ = 1.5406Å) operated at 40 kV and 40 mA.X-ray photoelectron spectroscopy (XPS) data were obtained with an AMICUS/ESCA 3400 electron spectrometer from Kratos Analytical (Manchester, UK) using Mg Kα radiation (20 mA, 12 kV).Binding energies were referenced to the C 1s line at 284.8 eV for adventitious carbon.We conducted H 2 temperature programmed reduction (H 2 -TPR) testing to analyze the redox properties of the catalysts using a Micromeritics ChemiSorb 2720TPx system (Micromeritics Instrument Ltd., Norcross, GA, USA) in a temperature range of 20 • C to 800 • C at a rate of 10 • C/min with a gas (10 vol% H 2 relative to Ar) flow rate of 40 mL/min, and then retained for 20 min at 800 • C.

Activity Measurement
We prepared MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalytic activity studies using a fixed bed microreactor.The reactor was composed of a stainless steel tube with a 10.0 mm inner diameter.Prior to conducting the experiments, quartz sand and quartz wool were placed inside the reaction tube to ensure contact between the powdered catalysts and the thermocouple.The typical composition of the simulated flue gas was 500 ppm NO (denoted as [NO] in ), 500 ppm NH 3 (denoted as [NH 3 ] in ), and 5 vol% O 2 with N 2 as the balance gas.The total volume flow was 100 mL/min, representing a GHSV of 30,000 h −1 .The catalytic activity of the powdered catalysts was evaluated in the temperature range of 20 • C to 400 • C during testing according to the exiting concentrations of NO (denoted as [NO] out ) and NH 3 (denoted as [NH 3 ] out ) determined by Fourier transform infrared (FTIR) spectroscopy (Nicolet IS10, Thermo Fisher Scientific, Waltham, MA, USA).The NO conversion ([NO] conversion ) and N 2 selectivity ([N 2 ] selectivity ) were calculated using the following equations: The catalytic performance of MnO x -Fe 2 O 3 /VMT MHCs was evaluated using a similar fixed bed microreactor composed of a quartz tube with a 5.0 cm inner diameter.The simulated flue gas was composed of 500 ppm NO and 500 ppm NH 3 with air as the balance gas.Prior to testing, each MHCs sample was placed in the reaction tube, and then exposed to the simulated flue gas for 1 h to eliminate the influence of adsorption on the catalysts.Then, [NO] out was measured online using a flue gas analyzer (QUINTOX-KM9106, Kane International, New York, NY, USA), and [NO] conversion was calculated using Equation (1).Unless otherwise stated here, all other conditions were equivalent to the conditions employed for powdered catalysts.

Conclusions
This paper presented the successful preparation of MnO x -Fe 2 O 3 /VMT catalysts with a layered structure by an impregnation method for the first time.The as-prepared MnO x -Fe 2 O 3 /VMT catalysts exhibited high NO conversion and N 2 selectivity for the NH 3 -SCR of NO in the temperature range of 20-200 • C. The catalysts provided an excellent NO conversion of 96.5% at 200 • C with GHSV = 30,000 h −1 and an NO concentration of 500 ppm.Compared with VMT and MnO x /VMT catalysts, the results of extensive characterization indicated that the high catalytic activity of the MnO x -Fe 2 O 3 /VMT catalysts can be attributed to a number of advantageous properties, such as a large specific surface area, high ratios of Mn 4+ /Mn total and Fe 3+ /Fe total , and easily reduced Mn species.In addition, the MnO x -Fe 2 O 3 /VMT powdered catalyst was successfully employed to form MHCs by an extrusion method.The as-prepared MnO x -Fe 2 O 3 /VMT MHCs provided an NO conversion of 98.6% at 150 • C with GHSV = 4000 h −1 .Moreover, the MHCs presented excellent De-NO x performance at low temperature, obtaining an NO conversion of 62.2% at 20 • C with GHSV = 1000 h −1 .Furthermore, the MnO x -Fe 2 O 3 /VMT MHCs also provided excellent cycling stability, and maintained comparable NO conversion values even after 10 h.Finally, the MnO x -Fe 2 O 3 /VMT MHCs demonstrated excellent SO 2 resistance at low temperature (particularly at room temperature).Therefore, the prepared MnO x -Fe 2 O 3 /VMT MHCs offer considerable potential for low or even room temperature De-NO x applications in stationary stack source emissions.

Figure 4 .
Figure 4. H 2 -TPR curves of the VMT support and MnO x /VMT and MnO x -Fe 2 O 3 /VMT catalyst samples.
x -Fe 2 O 3 /VMT MHCs provided an NO conversion of 79.1% at 50 • C. Surprisingly, the MHCs exhibited excellent NO conversion values of 62.2% and 50.2% with GHSV = 1000 h −1 and 2000 h −1 , respectively, at 20 • C. The excellent SCR performance obtained for the proposed MnO x -Fe 2 O 3 /VMT MHCs will strongly contribute to low-temperature De-NO x processes, and suggests that even room temperature processes are feasible.Catalysts 2018, 8, x FOR PEER REVIEW 9 of 14 conversion values of 62.2% and 50.2% with GHSV = 1000 h −1 and 2000 h −1 , respectively, at 20 °C.The excellent SCR performance obtained for the proposed MnOx-Fe2O3/VMT MHCs will strongly contribute to low-temperature De-NOx processes, and suggests that even room temperature processes are feasible.

3. 1 . 1 .
Preparation of MnO x /VMT and MnO x -Fe 2 O 3 /VMT Catalysts Chmielarz et al. investigated PCHs based on montmorillonite, saponite, and VMT modified with Fe or Cu as catalysts [31].Among these catalysts, PCHs involving VMT achieved the best SCR activity at 400 • C. Samojeden et al. developed VMT supports prepared by modifications with nitric acid, and the resulting catalysts formed by impregnation with Cu provided an NO conversion of 94.3% at 350 • C

Table 1 .
Physical properties of VMT support and catalyst samples.

Table 2 .
Surface compositions of representative catalyst samples obtained by XPS analysis.