**1. Introduction**

As one of the major air pollutants, nitrogen oxides (NOx) are considered to cause a series of environmental problems, such as acid rain, smog, and greenhouse effects [1]. The selective catalytic reduction (SCR) of NOx with NH3 (NH3-SCR) is one of the most effective techniques for reducing NOx from stationary resources caused by fossil fuel combustion (e.g., coal-burning power plants) [2]. In the past few years, considerable research work has been performed on the development of high efficiency catalysts for NH3-SCR [3]. V2O5-WO3-TiO2 catalysts have been widely employed as commercial catalysts; nevertheless, the disadvantages of these catalysts are their high operating temperatures (300-400 ◦C) and the toxicity of vanadium [4]. Hence, it is of grea<sup>t</sup> importance to find a new type of catalyst which is able to effectively remove NOx at low temperature since the catalyst is located behind the desulfurizer and electrostatic precipitator system to reduce the cost of NH3-SCR. A series of transition metal oxides such as MnOx, FeOx, CoOx, and CeOx supported on different carriers have been studied to raise low temperature activity [2,5,6]. Notably, manganese-oxide-based catalysts have shown promising catalytic activity among the studied catalysts. However, the relatively lower specific surface area of traditional carriers might hinder the further application of manganese-oxide-based catalysts. Thus, it is especially important to find a carrier substitute with a large specific surface area.

Among the potential candidates, metal–organic frameworks (MOFs) have attracted significant attention due to their large specific surface area, low density, high chemical tenability, and controlled structure. It has been reported that a Ni-MOF activated at 220 ◦C achieved a NO conversion efficiency of over 92% for a large operating-temperature range of 275 to 440 ◦C [7]. MOF-74 is a potential support for the low-temperature NH3-SCR process owing to its coordinatively unsaturated metal sites (CUSs), highly dispersed and absolute exposed metal sites, large specific surface areas, and high porosity. According to our previous work [8], the Mn-MOF-74 catalyst has good catalytic performance for the low-temperature reaction of NH3-SCR, and we found the NO conversion to be nearly 99% at 220 ◦C. However, the NO conversion of Mn-MOF-74 was observed to drop by 44% after the adding of 5% water, which hinders the further application of these materials. It has been reported that the presence of water could reduce the capacity of gas adsorption significantly and destroy the crystal structure of MOF-74 [9]. It is therefore necessary to improve the water stability of the prepared MOF materials under the premise of maintaining good denitrification performance.

At present, two major strategies are being employed to raise the water-resistance of MOFs to expand their applications. The most e ffective strategy for preparing MOFs with water stability is to introduce strong coordination bonds [10]; another method is to install a hydrophobic moiety around the coordination sites or on the working surface of the crystal to prevent corrosion from water molecules [11–13]. It has been proven that the latter is an e fficient way to enhance the water resistance of MOF materials. Wu and his co-authors enhanced the water resistance of IRMOF1 by adding water-repellent functional groups [12].

In this work, as-synthesized Mn-MOF-74 catalysts were modified via two methods to promote water resistance. One strategy used was to cover water-in-oil surfactants polyethylene oxidepolypropylene-polyethylene oxide (P123) or polyvinylpyrrolidone (PVP) on the surface of Mn-MOF-74 to increase the external surface hydrophobicity of MOFs, producing P123-Mn-MOF-74 and PVP-Mn-MOF-74. The other was to introduce hydrophobic groups (-CH3) to the ligand of Mn-MOF-74 to synthesize Mn-MOF-74-CH3 through a post-synthesis modification (PSM) method. In addition, the effects of these groups on crystal structure, morphology, and thermal stability were investigated by powder X-ray di ffraction (PXRD), FT-IR, SEM, TEM, and thermogravimetry mass spectrometry (TG-MS), et al. In addition, NH3-SCR performances with low temperature and water resistance for the prepared catalysts were studied.

#### **2. Results and Discussion**

#### *2.1. Characterization of MOFs*

Figure 1 shows the PXRD patterns of Mn-MOF-74, P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74-CH3. As can be seen from this figure, the two strongest peaks at 2θ = 6.5–7◦ and 11.5–12◦ of all the samples correspond to the crystal faces (2-10) and (300), respectively, which are characteristic di ffraction peaks of MOF-74. The relative strength of the two di ffraction peaks of P123-Mn-MOF-74 and PVP-Mn-MOF-74 accord with the standard pattern while the peak intensity of Mn-MOF-74-CH3 at 2θ = 11.5–12◦ is higher than that at 2θ = 6.5–7◦. The patterns of -CH3-functionalized Mn-MOF-74 are inconsistent with the standard pattern, which was probably caused by the synthesis method. In other words, the harsher preparation conditions of Mn-MOF-74-CH3 did have an influence on the structure of MOF-74, but the main crystal structure was maintained. In summary, the introduction of surfactants P123 and PVP and methyl had little e ffect on the crystal structure of MOF-74.

To further investigate the influence of modification on catalyst structure, FT-IR tests were carried out on Mn-MOF-74, P123/PVP-Mn-MOF-74, and Mn-MOF-74-CH3 (the spectra are displayed in Figure S1). The peak at 1625 cm<sup>−</sup><sup>1</sup> can be assigned to ν(-COO) of DHTP (2,5-dihydroxyterephthalic acid) [13]. The bands at 1558 cm<sup>−</sup><sup>1</sup> and 1400 cm<sup>−</sup><sup>1</sup> can be assigned to <sup>ν</sup>(C=C), which belongs to the skeleton vibration of the benzene ring [14,15] of DHTP. The peak at 1245 cm<sup>−</sup><sup>1</sup> can be assigned to ν(C-N) of DMF (N, N-Dimethyl formamide) [16,17], demonstrating that the solvent DMF exists on the catalysts' surface or in the channel of the catalysts. The bands at 888 cm<sup>−</sup><sup>1</sup> and 811 cm<sup>−</sup><sup>1</sup> can be attributed to a ν(C-H) oscillatory vibration in and out of the plane of the benzene ring [18]. The peaks at 579 cm<sup>−</sup><sup>1</sup> and 473 cm<sup>−</sup><sup>1</sup> belong to the vibration absorption ν(Mn-O) [19,20], which confirms the existence of a metal–ligand bond. Notably, as shown in Figure S1b, Mn-MOF-74-CH3 has a weak absorption peak at 2960 cm<sup>−</sup>1, which belongs to the stretching vibration absorption of ν(C-H) in alkanes [19], proving that


**Figure 1.** The powder X-ray diffraction (PXRD) patterns of Mn-MOF-74, polyethylene oxidepolypropylene-polyethylene oxide (P123)/polyvinylpyrrolidone (PVP)-Mn-MOF-74, and Mn-MOF-74-CH3.

As can be seen from Figure S2, the four samples showed spherical particles. The morphology of Mn-MOF-74-CH3 displayed almost no change except for there being a few tiny block crystals on the surface; the surfaces of P123-Mn-MOF-74 and PVP-Mn-MOF-74 were almost wrapped with a mass of tiny particles compared to Mn-MOF-74. The results sugges<sup>t</sup> that the original spherical morphology of Mn-MOF-74 remained unchanged whether the water-in-oil surfactants P123 and PVP were introduced or ligand methyl functionalization was performed.

The edge of each sample was observed using TEM characterization techniques and the results are shown in Figure 2. It may be noted that there is no coating on the edges of Mn-MOF-74 and Mn-MOF-74-CH3, while the edges of P123-Mn-MOF-74 and PVP-Mn-MOF-74 are obvious, which should be due to the coating of water-in-oil surfactants P123 and PVP, respectively. Compared with PVP-Mn-MOF-74, the P123-modified Mn-MOF-74 showed better surfactant dispersion and the particle size of the latter was more uniform. In summary, it can be speculated that the surfactants P123 and PVP were coated on the surface of the Mn-MOF-74 successfully.

**Figure 2.** TEM images of (**a**) Mn-MOF-74, (**b**) P123-Mn-MOF-74, (**c**) PVP-Mn-MOF-74, and (**d**) Mn-MOF-74-CH3. Illustration: edge of the related samples.

The prepared catalysts were analyzed by thermogravimetric analysis in air and under a nitrogen atmosphere and the results are shown in Figure S3a,b. Taking Mn-MOF-74-CH3 as an example, the ion fragments of decomposition products in the air atmosphere and N2 atmosphere were detected by mass spectrometry, as shown in Figure S3c,d.

From Figure S3a, it can be observed that the weight loss of the three catalysts showed two significant stages under air conditions. In the first stage, the mass loss ratio of P123-Mn-MOF-74 can be seen to be about 20% near 235 ◦C, which can be attributed to the removal of CHCl3, while that of PVP-Mn-MOF-74 and Mn-MOF-74-CH3 are about 20% at 275 ◦C, which could be ascribed to methanol and CHCl3 (as can be seen in Figure S3c), respectively. The second weight loss of all the samples is about 35% at about 320 ◦C, which is due to the total collapse of the skeleton structure. To sum up, it can be seen from the results that the three catalysts can maintain the integrity of the crystal structure when stored in air.

As shown in Figure S3b, there are three clear mass losses for the three catalysts in the N2 atmosphere. The first stage is about 10% near 105 ◦C for the Mn-MOF-74-CH3 sample, which can be attributed to the removal of CHCl3 only (no water, methanol, or DMF ion fragmentations can be detected from the MS results shown in Figure S3d), suggesting that solvent molecules in the tunnel were almost discharged in the pre-preparation evacuation process. For P123-Mn-MOF-74 and PVP-Mn-MOF-74, it is known from the preparation method that the mass loss can be ascribed to the removal of CHCl3 and methanol. The second stage is in the vicinity of 317 ◦C and is caused by the partial collapse of the skeleton structure from the MS results of Figure S3d. The third stage appears above 535 ◦C, suggesting the complete collapse of the MOF-74 skeleton structure. Combined with TG and MS results, the activation conditions of the samples should be set as N2 atmosphere, 200 ◦C for 3 h, and a heating rate of 2 ◦C/min.

#### *2.2. The Low-Temperature SCR Performance*

Low-temperature SCR catalytic performance was tested to investigate the effects of the surfactants P123 and PVP and ligand methyl functionalization on the catalytic performance of Mn-MOF-74 at low temperature. As shown in Figure 3, Mn-MOF-74 performed well with regard to low-temperature NH3-SCR activity, but the water resistance still needs further study. As the temperature increased, the NO conversion of all the samples exhibited a rising trend; when the temperature rose to 280 ◦C, the NO conversion of Mn-MOF-74-CH3 reached a maximum of 93.2%. The NO conversion of P123-Mn-MOF-74 and PVP-Mn-MOF-74 reached maxima of 92.1% and 71.8% at 265 ◦C and 250 ◦C, respectively, and then decreased with the continuous increase in temperature, which was caused by the collapse of the skeletal structure. Compared with the low-temperature SCR catalytic performance curve of Mn-MOF-74, the performance order was Mn-MOF-74 > Mn-MOF-74-CH3 > P123-Mn-MOF-74 > PVP-Mn-MOF-74. The surfactants P123 and PVP and the methyl ligands existing in the pores may even have been partially covered in the metal active center, hindering the approaches of reactant molecules and thus depressing the SCR catalytic performance.

**Figure 3.** Low-temperature selective catalytic reduction (SCR) activities of P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74-CH3 (gas flow rate: 100 mL/min; gas composition: NO 500 ppm, NH3 500 ppm, O2 5%, and Ar as balance gas).

#### *2.3. Stability and Water Resistance Study*

In this work, stability tests were carried out at temperatures corresponding to the maximum NO conversion of catalysts (MOF-74 240 ◦C, P123-Mn-MOF-74 250 ◦C, PVP-Mn-MOF-74 265 ◦C, and Mn-MOF-74-CH3 280 ◦C) for 12 h; the results are displayed in Figure 4. It was concluded that all the samples showed good stability at their own optimal temperature during the 12 h tests.

**Figure 4.** Stability test results of Mn-MOF-74, P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74- CH3 under an SCR atmosphere at their own optimal conditions (gas flow rate: 100 mL/min; gas composition: NO 500 ppm, NH3 500 ppm, O2 5%, H2O (g) 5%, and Ar as balance gas).

To study the water resistance of the prepared samples, 5 vol.% H2O was added into the feed gas and low-temperature catalytic activities were investigated (as shown in Figure 5). For a clearer explanation, the maximum NO conversion and the corresponding temperature, stability, and water resistance of Mn-MOF-74, P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74-CH3 have been summarized in Table 1.

**Figure 5.** Stability test results of Mn-MOF-74, P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74- CH3 under an SCR-H2O atmosphere at their own optimal conditions (gas flow rate: 100 mL/min; gas composition: NO 500 ppm, NH3 500 ppm, O2 5%, H2O (g) 5%, and Ar as balance gas).

According to the results given in Figure 5 and Table 1, the NO conversion of Mn-MOF-74 decreased by 44% after adding H2O. Combined with the SCR activity results, Mn-MOF-74 did show better catalytic activity, while the water resistance of it was very poor. The reason for this might have been that OH species of the dissolved H2O were bonded to the exposed CUSs, thereby reducing the number of active sites and reducing the activity [21]. After modification, the NO conversion of 71.8% for PVP-Mn-MOF-74 and 81.2% for Mn-MOF-74-CH3 can be observed to be higher than that of 55% for Mn-MOF-74 in the SCR-H2O atmosphere, and then be recovered to the original level

when removing H2O, suggesting that the water resistance of PVP-Mn-MOF-74 and Mn-MOF-74-CH3 increased greatly. Notably, the NO conversion of P123-modified Mn-MOF-74 remained at 92.1% after the addition of 5 vol.% H2O, which indicates that the introduction of P123 did enhance the water resistance of Mn-MOF-74.

**Table 1.** The effects of adding H2O on the low-temperature SCR catalytic activities of Mn-MOF-74, P123-Mn-MOF-74, PVP-Mn-MOF-74, and Mn-MOF-74-CH3.


#### **3. Materials and Methods**
