Porous Zeolitic Imidazolate Framework Loaded Mn as an Efficient Catalyst for the Selective Catalytic Reduction of NOx with NH3

Received: August 14, 2021
Revised: September 10, 2021
Accepted: September 14, 2021

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Cite this article:

Chen, Y., Yu, M., Wang, R. (2021). Porous Zeolitic Imidazolate Framework Loaded Mn as an Efficient Catalyst for the Selective Catalytic Reduction of NO_x with NH₃. Aerosol Air Qual. Res. 21, 210201. https://doi.org/10.4209/aaqr.210201

HIGHLIGHTS 

• [email protected] was synthesized with Mn species highly dispersed on ZIF-8.
• [email protected] showed an over 90% NO conversion at 225–400°C.
• Ideal activity and stability of [email protected] was prove after 12 hours' test.
• [email protected] exhibited high resistance to SO₂ without H₂O.
• The use of [email protected] as a novel catalyst for SCR reaction is feasible.

Keywords: Nitrogen oxides, SCR, Zeolitic imidazolate framework, ZIF-8

1 INTRODUCTION

Nitrogen oxides (NO_x), mainly NO and NO₂, is a kind of common gaseous pollutant, and the removal of NO_x has long been a difficult problem in air pollution control. The sources of NO_x can be divided into two categories, one is the mobile sources and the other is the fixed sources. The fixed sources mainly refer to the flue gas emitted from the combustion of fossil fuels such as coal and oil in the process of industrial production, and the mobile sources are generally thought to be the exhaust gas emitted from automobile engine. With the rapid development of modern industrialization, a large amount of fossil fuels have been exploited and consumed, making the NO_x emitted into the atmosphere far exceed the carrying capacity and self-purification capacity of the atmospheric environment. NO_x brought severe harm to environment, such as the photochemical smog, acid rain and ozone holes. All these environmental pollution phenomena are related to the emission of NO_x, and have a serious threat to our human beings. Consequently, it is essential to control the NO_x in the atmosphere (Brandenberger et al., 2008; Busca et al., 2008; Forzatti, 2001; Twigg, 2007).

NH₃-SCR has been proven to be the most effective technique for abatement of NO_x (Chen et al., 2016). And the catalyst with high denitrification efficiency is the key of the SCR technique. So far, the SCR catalysts can be divided into the following categories according to their active components: noble metal catalysts (Li et al., 2010; Qi et al., 2004), metal oxides catalysts (Qi et al., 2003a; Wei et al., 2018; Shen et al., 2016) and molecular sieve catalysts (Brandenberger et al., 2008; Delahay et al., 2005; Gao et al., 2017). Though the commercial catalysts V₂O₅-WO₃(MoO₃)/TiO₂ have been widely used, the defects like narrow temperature window and toxicity of vanadium pentoxide still exist (Yi et al., 2016). Hence, the novel catalysts with wide temperature window and environment-friendly properties is desirable.

Manganese oxide (MnO_x), a kind of transition metal oxide, has attracted great attentions due to its various labile oxygen species and high activity in NH₃-SCR reaction (Wu et al., 2007). Min et al. (2007) prepared different types of MnO_x catalysts through a precipitation method, and got a high NO_x conversion for the SCR of NO_x with NH₃. To enhance the performance of MnO_x catalyst, a series of doping species and supporter was applied. Peng et al. (2016) used Eu to modify MnO_x catalyst to promote SCR activity, and the results showed that MnEuO_x exhibited a high NO_x conversion in a wide temperature range of 150–400°C. CeO₂ was a another metal species used to improve the SCR activity of MnO_x. Qi et al. (2003b, 2004) prepared mixed manganese and cerium oxides, and the MnO_x-CeO₂ was highly active in SCR reaction obtaining almost 95% NO conversion at 150°C. Singoredjo et al. (2010) prepared the alumina supported MnO_x exhibiting a high NO_x conversion during 120–300°C. TiO₂ is another kind of support commonly used to facilitate the SCR activity of MnO_x, Park et al. (2013) loaded Mn on synthesized TiO₂ through impregnation method, exhibiting a high NO conversion of almost 100% at 150°C. Besides, a series of metal species such as Fe, Ce, V, Sm, Zr, were used to modify the MnO_x/TiO₂ to avoid poor resistance to SO₂ of MnO_x. Yet, the result could not meet the requirements of industrialization, further investigation is still needed (Yang et al., 2016; Niu et al., 2016; Zhang et al., 2018; Sun et al., 2018).

Metal organic frameworks (MOFs), the porous crystalline materials, have drawn extensive research interests due to their higher specific surface area, higher porosity, and more stable porous structure compared to the metal oxides (Lee et al., 2009). It is advantageous for them to adsorb the reactants and provide great gas storage, separation and catalysis, besides, the high specific surface, well ordered porous structure, and regular crystal structure have proven the MOFs to be ideal support (Zhang et al., 2016; Wang et al., 2016; Zhang et al., 2017). Zhang et al. (2017) loaded Mn and Ce onto MOFs through in situ and impregnation methods, and studied their catalytic activities in SCR reaction, the results showed that [email protected] had a high NO_x conversion in a wide temperature range. Wang et al. (2016) prepared the CeO₂/MIL-100(Fe) catalysts by encapsulating ceria nanoparticles into MIL-100(Fe) through impregnation, the CeO₂/MIL-100(Fe) catalyst showed a high NH₃-SCR activity and great resistance to SO₂ at 196–300°C. Zeolitic imidazolate framework (ZIFs), a sort of MOFs, has higher specific surface area and better thermal stability than ordinary MOFs (Huang et al., 2010; Park et al., 2006). Among them, ZIF-8 is the most widely studied ZIFs with a specific surface area of 1400 m² g^–1 and great thermal stability (Park et al., 2006; Venna et al., 2010), and has been widely used in gas adsorption, hydrogen storage and catalysis. However, few researchers have applied ZIF-8 in NH₃-SCR reaction (Banerjee et al., 2008; Küsgens et al., 2009; Nguyen et al., 2012).

In this work, ZIF-8 was synthesized through a solvothermal method, and MnO_x was firstly loaded on ZIF-8 by impregnation. The activity of the catalysts in NH₃-SCR reaction, the stability of the catalysts, the effect of the addition amount of Mn species on catalytic activity as well as the resistance to H₂O and SO₂ were investigated. Besides, the characterization of the catalysts including XRD, SEM, EDS Mapping, XPS, BET, and H₂-TPR, TGA was carried out to research the catalysts for further information.

 
2 EXPERIMENTAL

 
2.1 Materials

All the chemicals were available commercially and used without further purification. 2-methylimidazole (2-HMelM, 99%) Zinc nitrate Hexahydrate (Zn(NO₃)₂·6H₂O, 99%) were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd. N,N-Dimethylformamide (DMF 99.5%) and 50% Manganese nitrate water solution (Mn(NO₃)₂ 50%) were purchased from Tianjin Fuchen Chemical reagents factory.

 
2.2 Synthesis of Catalyst

2.2.1 Synthesis of ZIF-8 particles

ZIF-8 was synthesized by solvothermal method according to the following procedure (Gee et al., 2013; Tran et al., 2011): A solid mixture of Zn(NO₃)₂·6H₂O (0.717 g, 2.41 mmol) and 2-methylimidazole (H-MeIM) (0.18 g, 2.19 mmol) was dissolved in 50 mL DMF, the solution was transferred into a teflon-line stainless steel autoclave, and heated to 140°C at a rate of 5°C min^–1 in a temperature-programmed oven and maintained for 24 h. After cooling to ambient temperature, the crystal was collected and washed with DMF for three times. Finally, these samples were dried at 60°C overnight to remove the residual solvents.

 
2.2.2 Synthesis of [email protected]

The Mn was loaded on ZIF-8 via impregnation method reported by Zhang et al. (2017). Adding 0.4, 0.6, 0.8, and 1 mL of Mn(NO₃)₂ respectively into 20 mL methanol to get Mn solutions, and dispersing 1 g of ZIF-8 into 30 mL methanol under ultrasonication for 15 min to get a ZIF-8 solution. Then, the Mn solution was slowly dropped into ZIF-8 solution under vigorous stirring for 4 h, followed by filtration to get the precipitate, after that, the precipitate was dried overnight at 60°C. Finally, the precipitate was heated at 300°C for 5 h. The obtained samples were denoted as [email protected](0.4), [email protected](0.6), [email protected](0.8) and [email protected](1), respectively. For convenience, the [email protected] without note represents the [email protected](0.8).

 
2.3 Characterizations of the Catalysts

The XRD patterns were obtained on a Bruker D8 diffractometer using Cu Kα radiation between 2θ = 5° and 45°. The micro-morphology of [email protected] was investigated by scanning electron microscope (SEM) (SU 8010), and energy dispersive spectroscopy (EDS) mapping images were obtained on the same device. The X-ray photoelectron spectroscopy (XPS) analysis was carried on a Thermo Fisher Scientific ESCALAB 250 spectrometer, the correction of binding energy shift was referenced to C 1s line at 284.6 eV, and the spectra of the Mn 2p and O 1s were recorded. The specific surface area analysis was conducted on Micromeritics Tri StarII 3020 Surface Arae and Porosity Analyzer with a N₂ adsorption, and the results were calculated with the multi-point Brunauer-Emmett-Teller (BET) approach. The thermogravimetric analysis (TGA) was carried on a TGA SDT Q600 thermogravimetric analyzer with a heating rate of 10°C min^–1 from 25°C to 600°C under nitrogen atmosphere. The hydrogen temperature programmed reduction (H₂-TPR) analysis was performed in which 100 mg of the samples were preheated to 450°C with a ramp rate of 10°C min^–1 under N₂ atmosphere. After cooling to the ambient temperature, the samples were heated up to 800°C at a ramp rate of 8°C min^–1 with a flow of H₂, and the consumption of H₂ was continuously recorded using the TCD detector.

 
2.4 Activity Tests

The tests of NH₃-SCR activity were carried out in a fix-bed quartz reactor (8 mm in diameter) heated by a tube furnace. The simulated flue gas contained 500 ppm NO, 500 ppm NH₃, 5% O₂, 3% H₂O (when used), 200 ppm SO₂ (when used) and balance N₂, which were regulated by mass flow controllers respectively. In each run of SCR test, 0.2 g of [email protected] was used and the total flow rate was controlled at 100 mL min^–1 corresponding to a gas hourly space velocity (GHSV) of approximately 240000 h^–1. The catalyst was heated to the temperature ranging from 100°C to 450°C at a rate of 10°C min^–1, and held at each temperature spots for 50 min. Concentrations of NO and NO₂ were simultaneously monitored by the NO and NO₂ analyzers (TH-9905), and N₂O in the outlet gas was measured by an Antaris™ IGS Gas Analyzer from Thermo Fisher Scientific Inc. The NO_x conversion rate and the N₂ selectivity were calculated as follows:

where the subscripts “in” and “out” refers to the inlet concentration and outlet concentration at steady state, respectively.

 
3 RESULTS AND DISCUSSION

 
3.1 Characterizations of the Catalysts

3.1.1 XRD

To determine the crystal structure of the catalysts, the XRD analysis of the synthesized ZIF-8 and [email protected] was carried out. It could be easily seen from Fig. 1 that the characteristic peak of synthesized ZIF-8 was in accordance with the simulated ZIF-8, which approved the ZIF-8 was successfully synthesized. As for the XRD pattern of [email protected], no apparent difference was found after the impregnation, indicating that the crystalline structure of ZIF-8 retained well. Yet, the characteristic peaks of Mn species were not observed which may be interpreted by that the Mn species were in amorphous phase (Li et al., 2017).

Fig. 1. XRD patterns of the catalysts.

 
3.1.2 SEM and EDS Mapping

SEM was carried to investigate the structure of [email protected]. It can be seen from Fig. 2 that the synthesized catalysts still retained its dodecahedral shape, which matched well with the results of XRD analysis. Moreover, EDS Mapping was performed to demonstrate the distribution of Mn, Zn species, and it could be seen that the elements of Mn and Zn are uniformly distributed on the whole surface of the catalysts, which was beneficial for adsorption and activation of the reactants, and was conducive to the NH₃-SCR catalytic activity.

Fig. 2. SEM image and EDS Mapping of [email protected].

 
3.1.3 XPS spectra

To further investigate the chemical state of Mn elements in the catalysts, XPS spectra were recorded and displayed in Fig. 3(a), from which it could be seen obviously that two prominent peaks at 642.89, 654.73 eV accompanied by two satellite peaks at 641.53, 653.46 eV were gained after fitting by Gaussian components. The peaks at 641.53 eV, 653.46 eV could be attributed to Mn²⁺ species, while the peaks at 642.89 eV and 654.73eV could be assigned to Mn³⁺ (Becerra et al., 2011; Guo et al., 2016). According to previous reports, the co-existence of Mn²⁺ and Mn³⁺ on the catalysts could facilitate the formation of oxygen vacancies, which played an important role in low temperature oxidation property (Liu et al., 2012). Besides, the high valence state of Mn³⁺ species would improve the reducibility of the sample (Cheng et al., 2017).

Fig. 3. XPS spectra of (a) Mn 2p and (b) O 1s for [email protected].

The O 1s XPS spectra of the [email protected] were shown in Fig. 3(b), two asymmetric peaks could be observed, indicating there were two distinct types of oxygen species. The lower binding energy peak at 530.27 eV was ascribed to the lattice oxygen species O^2– in the catalysts, and the sub-bands at 531.71 ev was attributed to surface chemisorbed oxygen species, such as O^2– or O^– (Atribak et al., 2011).

3.1.4 H₂-TPR

To investigate the redox behavior of [email protected], the H₂-TPR analysis was performed, and the results were presented in Fig. 4, from which it could be seen that there are two main reduction peaks. The first peak located at around 250°C could be assigned to the reduction of MnO₂ to Mn₂O₃ and the following peak around 520°C could be attributed to the further reduction of Mn₂O₃ to MnO (Zhao et al., 2016; Du et al., 2018). Normally, the operational temperature should be lower than 450°C. This prominent peak could be caused by the reduction of Mn³⁺ → Mn²⁺and the decomposition of ZIFs based on the TGA curves. Then, it could be speculated that the doping of Mn species prompted the catalysts to have the stronger redox behavior and oxygen storage capacity of the catalyst, which played an important role in the SCR activities (Li et al., 2016).

Fig. 4. H₂-TPR profiles of the catalysts.

 
3.1.5 BET analysis

The BET surface area of a catalyst support plays a significant role in the NH₃-SCR reaction. Besides, the high BET surface area of the support can improve the catalytic activity considerably by means of enhancing the dispersion of active components to avoid aggregation, as well as the transfer of charge (Han et al., 2015; Wang et al., 2013). The BET surface areas of all the samples were listed in Table 1. It can be seen that the high surface area of ZIF-8 would be beneficial to the high NO_x conversion and the surface area of the samples was increased with the increase of the addition of Mn species, however, when the addition amount of Mn species reached 1 mL (in Mn(NO₃)₂ solution volume), the surface area began to decline. The decrease in specific surface area may be due to the excessive Mn loading caused formation of large solids and pores clogging of the ZIFs materials. The variation trend of the surface area was in accordance with the SCR activities, which indicated that the BET surface area was an important factor that influenced the SCR activities of the samples.

 
3.1.6 Thermogravimetric analysis of [email protected]

As was well known, the MOFs usually have relatively poor thermal stability, yet, the thermal stability of the catalysts has a great impact on the SCR reaction. To investigate the stability of [email protected], the thermal gravimetric analysis was conducted, and the result shown in Fig. 5 exhibited a slightly weight loss of 5% from room temperature to 450°C, corresponding to the depletion of guest molecules, mainly H₂O, indicating the structure of [email protected] maintained well under 450°C, and [email protected] has a high thermal stability. When the temperature was higher than 450°C, a sharp weight loss of the sample appeared, which could be ascribed to the oxidation of organic ligands and the collapse of the structure of [email protected], which may account for the decrease in activity of the catalysts.

Fig. 5. The TGA curves obtained for the [email protected].

 
3.2 The Effect of Loading Amount of Mn Species

The NO_x conversion of all the catalysts in the range of 200–450°C was shown in Fig. 6, from which it could be seen that more Mn species leaded to higher catalytic activity of the samples till the [email protected](0.8) which presented the highest deNO_x efficiency compared with the rest Mn loadings. The NO_x conversion of [email protected](0.8) was over 90% in the range of 225–400°C, and reached 100% at 350°C. However, when the loading amount of Mn species exceeded 1 mL (in Mn(NO₃)₂ solution volume), the catalytic activity began to decline, which may be caused by the severe decrease of BET surface area. On the other hand, the NO_x conversion of all the samples increased with the rising of temperature within 200–350°C, and when the temperature went up to 400°C, the NO_x conversion began to decrease, which may be inferred from the TGA analysis that this decreased activity may be caused by the collapse of ZIF-8 framework.

Fig. 6. (a) Catalytic SCR activities and (b) N₂ selectivity of the catalysts. Reaction conditions: [NH₃] =500 ppm, [NO] =500 ppm, [O₂] = 3%, N₂ = balance and total flow rate =100 mL min^–1.

N₂O is also an important pollutant in the air. Using traditional Mn-based catalysts, a large amount of N₂O was produced by side reaction and therefore limited its application. In view of this, the formation of N₂O in the SCR reaction was tested, and the N₂ selectivity of catalysts with different loadings was studied. As shown in Fig. 6(b), the N₂O was formed in all catalysts. With the increase of temperature, ammonia was oxidized to N₂O gradually, and the nitrogen selectivity of the catalyst decreased continuously. Besides, the nitrogen selectivity of the catalyst decreased when the loading amount of Mn species increased. Yet the nitrogen selectivity was still over 70% with higher catalyst activity in the temperature window of 250–350°C.

3.3 Stability Test of [email protected]

Besides, the stability tests of the catalysts at 350°C was carried out, and as illustrated in Fig. 7, the NO_x conversion of [email protected] maintained high level after 12 h, presenting a good stability under the identical temperature, which may be ascribed to the porous structures of ZIF-8 and the strong interaction between Mn species and ZIF-8 (Zhang et al., 2014).

Fig. 7. The catalytic activity of [email protected]. Reaction conditions: [NH₃] =500 ppm, [NO] =500 ppm, [O₂] = 5%, N₂ = balance and total flowrate =100 mL min^–1, T = 350°C.

 
3.4 Effect of SO₂ and H₂O on [email protected]

As the existence of H₂O and SO₂ is unavoidable in the exhaust, even after desulfurization, the residual H₂O and SO₂ still exert a notable impact on SCR performance. Therefore, the effects of SO₂ and H₂O on the NO_x conversion of [email protected] were investigated. The results shown in Fig. 8 revealed that the NO_x conversion decreased slightly after introducing 200 ppm of SO₂, and then reached a relatively steady state. After cutting off the input of SO₂, the conversion of NO_x was gradually restored its original level and remained stable. Hence, it could be concluded that the [email protected] had a great resistance to SO₂, and the process of SO₂ inhibition is reversible. Besides, the negative effect of SO₂ on SCR reaction could be attributed to the following aspects: on one hand, the SO₂ would compete with the reactants for the active sites, on the other hand, the formed sulfate depositing (ammonium sulfate and ammonium bisulfate species) would block the active sites (Lu et al., 2015). Yu et al. (2010) proposed that the porous structure was in favor of the great SO₂ resistance. The high sulfur resistance of Mn based catalysts was related to porous silica support (Huang et al., 2008). Thus, it could be inferred that the great SO₂ resistance of [email protected] may be ascribed to the porous structure of ZIF-8.

Fig. 8. The effect of SO₂ on the SCR activities of [email protected]. Reaction conditions: [NH₃] = 500 ppm, [NO] = 500 ppm, [SO₂] =200 ppm, [O₂] = 5%, N₂ = balance and total flowrate =100 mL min^–1, T = 350°C.

The impact of H₂O and SO₂ was also investigated. The NO_x conversion decreased to 65% under the coexistence of H₂O and SO₂ and could not recover to the original value after eliminating the SO₂+H₂O feeding as shown in Fig. 9, which demonstrated that there was a synergistic inhibition effect between H₂O and SO₂, and massive sulfate species and deposition were formed and blocked the active sites.

Fig. 9. The effect of SO₂ and H₂O on the SCR activities of [email protected]. Reaction conditions: [NH₃] = 500 ppm, [NO] = 500 ppm, [SO₂] = 500 ppm, [O₂] = 5%, [H₂O] = 3%, N₂ = balance and total flowrate = 100 mL min^–1, T = 350°C.

 
4 CONCLUSIONS

The XRD patterns and SEM image demonstrated that the [email protected] was synthesized successfully, and the Mn species were highly dispersed on ZIF-8. The Mn 2p XPS spectra proved that the Mn²⁺ and Mn³⁺ coexisted on the catalysts and played an important role in redox process.

The prepared [email protected] showed an over 90% NO conversion at the temperatures ranging from 225–400°C, the excellent catalytic activity may attributed to the huge specific surface area of ZIF-8, as the variation trend of the surface area was in accordance with the SCR activities. Besides, the activity and stability for [email protected] was proved to be ideal after 12 h. Furthermore, the [email protected] exhibited strong tolerance against SO₂, which may be ascribed to the porous structure of ZIF-8. It is noteworthy that the catalytic activity of [email protected] was inhibited considerably in the presence of H₂O and SO₂. As a whole, this study demonstrated that the ZIF-8 is a kind of promising support and the use of [email protected] as a novel catalyst for SCR reaction is feasible.

 
CONFLICTS OF INTEREST

There are no conflicts to declare.

ACKNOWLEDGEMENTS

We would like to extend our sincere appreciation to the support from the National Natural Science Foundation of China (22078176).