Porous metal–organic framework MIL-100(Fe) as an efficient catalyst for the selective catalytic reduction of NO_x with NH₃

Abstract

The development of an efficient catalyst with excellent catalytic activity and high SO₂ resistance at low temperatures (<300 °C) remains a challenge for the selective catalytic reduction (SCR) reaction. In this study, we report that MIL-100(Fe) is an alternative catalyst for the SCR of NO_x with NH₃, and it exhibits higher NO_x conversion than the conventional V₂O₅–WO₃/TiO₂ catalyst below 300 °C. In addition, the effect of H₂O and SO₂ on the catalytic activity is reversible, and NO_x conversions are recovered after ceasing the flow of SO₂ and H₂O. In situ diffuse reflectance infrared Fourier transform spectroscopy results indicate that both ionic NH₄⁺ and coordinated NH₃ existed on MIL-100(Fe) and that the reaction primarily followed the Langmuir–Hinshelwood mechanism, in which NH₄⁺ reacts with NO₂ formed from NO oxidation over iron sites. Furthermore, the redox properties of the iron species (Fe³⁺ + e⁻ ↔ Fe²⁺) could play a significant role in activating the reactants.

---

1. Introduction

The selective catalytic reduction (SCR) of NO_x with NH₃ is an efficient method for removing NO_x from stationary sources.^1–3 In practice, the most widely used catalyst for this process is V₂O₅–WO₃(MoO₃)/TiO₂, which is effective at high operating temperatures (300–400 °C).³ Therefore, the SCR unit must be located upstream from the electrostatic precipitator. However, this placement accelerates the deactivation of the catalyst due to the high concentrations of SO₂ and ash. In addition, this type of catalyst is still not satisfactory due to the volatility and toxicity of vanadium as well as the high rate of conversion of SO₂ to SO₃.⁴ Therefore, substantial research has been devoted to the development of new vanadium-free catalysts for low temperature SCR (<300 °C) such that the SCR unit can be located downstream of the electrostatic precipitator and desulfurizer.

At present, the development of efficient SCR catalysts especially low temperature catalysts causes extensive attention. The transition metal oxide supported catalysts or combined catalysts as Ce-based catalyst, Cu-based catalysts, and Mn-based catalysts displayed good catalytic activity compared with the conventional V₂O₅–WO₃/TiO₂ catalysts in the SCR process.^5–10 However, the catalysts exhibit poor SO₂ tolerance, especially at relatively low temperature.¹¹ Therefore, a catalyst with both high activity and high sulfur resistance must be developed.

Metal–organic frameworks (MOFs), which are known to be porous crystalline materials with a large surface area, have recently attracted increasing attention as heterogeneous catalysts due to their unique properties (i.e., high metal content and porosity).¹² MOFs have an infinite network structure constructed of multitopic organic ligands and metal ions, and the pore structure can be customized using the appropriate linkers.¹³ The high degree of tunability along with a high surface area and pore topology indicate that MOFs with transition metal centers might serve as potential catalysts for the SCR reaction, especially in the presence of SO₂, because the large surface area and porosity are beneficial for dispersing the catalytic components and reducing the accumulation of sulfates on the surface to improve catalytic activity and SO₂ resistance.¹⁴ Another important feature that allows MOFs to serve as SCR catalysts is the materials' high metal content; moreover, in MOFs, all of the active sites are ultra-highly dispersed and homogeneous due to the highly crystalline nature of the materials.^15–17 Active sites are freely accessible to reagent molecules. In addition, because the metal ions in MOFs are bonded to organic linkers, the combination of metal cations and organic linkers prevents SO₂ poisoning of the active sites.¹⁸ Therefore, MOFs with transition metals are anticipated to be efficient and sulfur-resistant catalysts for the SCR.

Iron-based catalysts, such as Fe oxides or Fe³⁺ exchange zeolites, are active and environmentally friendly for the NH₃-SCR due to their good activity and N₂ selectivity.^19–21 Therefore, MOFs constructed with iron centers could be an alternative SCR catalyst that exhibits high catalytic activity and high SO₂ resistance. MIL-100(Fe) (MIL: Materials of Institute Lavoisier) is a three-dimensional iron(III) trimesate that is composed of trimers of iron octahedra sharing a common vertex μ₃-O; these trimers are linked by benzene-1,3,5-tricarboxylate moieties to form hybrid supertetrahedra.²² MIL-100(Fe) contains a significant amount of Fe^III CUS (coordinatively unsaturated metal sites) and Fe^II CUS, which are due to thermal activation in an inert gas flow, that act as Lewis acid sites.²³ These sites facilitate the adsorption of reactants and promote catalysis. In addition, MIL-100(Fe) could kept the stable structure in the condition with the existence of NO_x, NH₃, SO₂, H₂O and CO₂.^24–27 Therefore, MIL-100(Fe) could be a stable catalyst in the SCR reaction condition.^24–26

In this study, we investigated the feasibility and performance of MOFs as a novel class of catalytic materials for the SCR of NO_x with NH₃ using the MIL-100(Fe) catalyst. Water and sulfur-resistance experiments were also performed to evaluate the durability of the catalyst in the presence of H₂O and SO₂. In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was employed to explore the mechanism of NO_x removal. To the best of our knowledge, there are no published reports describing the application of MOF materials for the SCR. This work may provide new insight into the design of efficient SCR catalysts with improved activity.

2. Experimental

2.1 Catalyst preparation

MIL-100(Fe) was prepared from the hydrothermal reaction of trimesic acid with metallic iron, HF, nitric acid and H₂O.²² The molar composition of the reaction mixture was Fe⁰ : 1,3,5-BTC : HF : HNO₃ : H₂O = 1.0 : 0.67 : 2.0 : 0.6 : 277 (1,3,5-BTC = 1,3,5-benzenetricarboxylic or trimesic acid). The reactant mixture was loaded in a Teflon autoclave and maintained at 150 °C for 12 h. The as-synthesized MIL-100(Fe) was further purified by a two-step treatment using hot deionized water (80 °C) and ethanol (60 °C) for 3 h. Finally, the resulting light-orange solid was dried at 100 °C overnight.

The conventional catalyst (i.e., V₂O₅–WO₃/TiO₂ (V: ∼1%, W: ∼9%)) was prepared using the incipient wetness method.² In a typical preparation, WO₃ was added to a TiO₂ support by impregnation with an aqueous solution of ammonium paratungstate complexed with citric acid, followed by drying and calcination at 550 °C for 3 h. Then, vanadium was added to the WO₃/TiO₂ system by impregnation starting from a water solution of ammonium metavanadate and oxalic acid, followed by drying and calcination at 550 °C for 2 h.

2.2 Characterization

Powder X-ray diffraction (XRD) patterns were recorded over the range of 3–40° on a Panalytical/Empyrean X-ray diffractometer with Cu Kα radiation. Infrared reflectance (IR) spectrums were recorded on a Bruker VERTEX 70 FTIR spectrometer equipped with DTGS detector in the range of 4000–600 cm⁻¹ with the resolution of 4 cm⁻¹. The morphology of MIL-100(Fe) was observed using a field-emission scanning electron microscope (SEM, Hitachi S-4800) operated at an accelerating voltage of 10 kV. The specific surface area was determined from a nitrogen adsorption isotherm using the Brunauer–Emmett–Teller (BET) model in Quantachrome SI. Thermogravimetric (TG) analysis was performed using a TG/DTA 6300 thermogravimetric analyzer from Seiko under an air stream (100 ml min⁻¹) at a heating rate of 5 °C min⁻¹ from 25 to 600 °C. X-ray photoelectron spectroscopy (XPS) was conducted on the surface analysis system (Thermal ESCALAB 250) using Al Kα radiation. The C 1s peak at 286.4 eV was used for the binding energy calibration.

In situ DRIFTS measurements were performed on a Bruker VERTEX 70 FTIR spectrometer equipped with a MCT detector cooled by liquid nitrogen. An approximately 10 mg sample was finely ground and pressed into a self-supported wafer. The total gas flow was maintained at 100 ml min⁻¹ by mass flow controllers. Prior to each experiment, the sample was pretreated at 250 °C for 10 h and cooled to room temperature in a flow of N₂. At each temperature, the background spectrum was recorded in flowing N₂ and automatically subtracted from the sample spectrum obtained at the same temperature. In the experiment, DRIFT spectra were recorded by accumulating 16 scans at a resolution of 4 cm⁻¹.

2.3 Measurement of catalytic activity

SCR reaction tests were performed in a fixed-bed quartz tube reactor (ϕ = 9 mm) containing approximately 0.25 g of MIL-100(Fe) and 0.46 g of V₂O₅–WO₃/TiO₂. The MIL-100(Fe) catalyst was pretreated in N₂ at 250°C for 10 h prior to the activity measurement. The feed gas consisted of 500 ppm NO, 500 ppm NH₃, 4% O₂, 500 ppm SO₂ (when used), 5% H₂O (when used), and N₂ balance. The total flow rate was 315 ml min⁻¹, and the gas hourly space velocity (GHSV) was 30 000 h⁻¹. The inlet and outlet NO_x (NO plus NO₂) concentrations were analyzed using a flow gas analyzer (ecom-J2KN, rbr Messtechnik GmbH Inc.). The concentration of NH₃ was recorded through by NH₃ analyzer (GXH-1050, Beijing), and the N₂O concentration was monitored by a gas chromatograph (Shimadzu GC-14C) with a Porapak Q column. NO_x conversion and N₂ selectivity were calculated as follows:

The conversion rate was calculated from the conversion of NO_x by using the pseudo-first-order kinetic equation as below.²⁷

r = −F/W × ln(1 − x) (mol NO_x g cat⁻¹ s⁻¹) (3)

In the equation, F is the flow rate of NO (mol s⁻¹), W is the weight of the catalyst (g) and x is NO_x conversion (%). The conversion rate is affected by the reactant concentration, and in this equation the reaction orders of the reactant concentration with respect to NO, NH₃ and H₂O were first order, zero order and zero order, respectively. And the effect of O₂ on the reaction rate is neglected due to its excessive concentration in the reaction.

3. Results and discussion

3.1 Physical and textural properties of MIL-100(Fe) catalyst

MIL-100(Fe) crystallizes in the cubic space group Fd3̄m (no. 227) with the cell parameter of a = 73.340 Å and V = 394 481.1 ų.²² Rietveld refinement plot for the XRD pattern of the as-synthesized MIL-100(Fe) was obtained by using the GSAS software.²⁸ The results demonstrated that the peak positions matched well with that of the simulated diffraction pattern of MIL-100(Fe) in the range of 3 to 40 degree, which indicated that the crystalline state of the sample was consistent with MIL-100(Fe) (ESI, Fig. S1†). The SEM image indicated the size of the octahedral crystal, which was approximately 1.5 μm, as shown in Fig. 1a. The specific surface area and pore volumes determined by N₂ physisorption were 2032 m² g⁻¹ and 1.12 cm³ g⁻¹, respectively, greater than those of the conventional metal oxide catalyst. Therefore, MIL-100(Fe) could provide a high surface area, which is beneficial for the SCR reaction.²⁹ In addition, the thermal stability of MIL-100(Fe) was studied by TG analysis from 25 to 600 °C in air, and the results are shown in Fig. S2.† A weight loss of 14.5% below 200 °C was due to the removal of physically and chemically adsorbed water.³⁰ The loss observed at 200–300 °C was only 2.7%, which indicated that MIL-100(Fe) was thermally stable blow 300 °C. This thermal stability could ensure the applicability of the MIL-100(Fe) catalyst over this temperature range.

Fig. 1 SEM image of MIL-100(Fe). (a) MIL-100(Fe) before NH₃-SCR activity test; (b) MIL-100(Fe) after 20 h NH₃-SCR activity test at 250 °C; (c) MIL-100(Fe) after NH₃-SCR durability test of resistance of SO₂ and H₂O at 250 °C.

3.2 Catalytic performance

3.2.1 NH₃-SCR performance. Fig. 2a shows the NO_x conversions of the NH₃-SCR reaction as a function of temperature over the MIL-100(Fe) and conventional V₂O₅–WO₃/TiO₂ catalysts. The NO_x conversion over MIL-100(Fe) rapidly increased when the temperature was higher than 150 °C, and a NO_x conversion of more than 97% was obtained at 245–300 °C. However, the NO_x conversion was only 51–90% within this temperature range over the V₂O₅–WO₃/TiO₂ catalyst. As shown in Fig. 2, more than 90% NO_x conversion for the V₂O₅–WO₃/TiO₂ catalyst was achieved above 300 °C. These results indicated that the MIL-100(Fe) catalyst was more active than the conventional V₂O₅–WO₃/TiO₂ catalyst for the SCR of NO_x at lower temperatures.

Fig. 2 NO_x conversions over MIL-100(Fe) and V₂O₅–WO₃/TiO₂. Reaction condition: 500 ppm NO, 500 ppm NH₃, 4% O₂ and N₂ balance, GHSV = 30 000 h⁻¹.

Compared with the conventional catalysts reported in the ref. 31 and 32 the conversion rate of 0.74 × 10⁻⁶ mol NO_x g cat⁻¹ s⁻¹ (calculated by eqn (3)) over MIL-100(Fe) at 200 °C is greatly higher than those of 0.09 × 10⁻⁶ mol NO_x g cat⁻¹ s⁻¹ and 0.13 × 10⁻⁶ mol NO_x g cat⁻¹ s⁻¹ over V₂O₅–WO₃/TiO₂ catalysts, which indicated MIL-100(Fe) catalyst was more efficient than that of the V₂O₅–WO₃/TiO₂ catalyst in the SCR reaction (Table 1).

Table 1 Comparison of SCR performance of various catalysts

Catalyst	Temperature (°C)	Rate (mol NOx g cat^−1 s^−1, 10^−6)
a The content of V2O5 is 1 wt%.b Commercial catalyst.
MIL-100(Fe)	200	0.74
V2O5–WO3/TiO2	200	0.09 (in this work)
V2O5–WO3/TiO2^a^,^24	200	0.09
V2O5–WO3/TiO2^b^,^25	200	0.13

---

However, the catalytic activity of MIL-100(Fe) decreased when the temperature was higher than 300 °C. The TG analysis results indicated that MIL-100(Fe) began to decompose, and the weight loss increased from 17.2 to 67.5% when the temperature increased from 300 to 405 °C. SEM images also indicated that MIL-100(Fe) collapsed after the SCR reaction at 325 and 350 °C (as shown in Fig. S3†). The XRD patterns of the samples used are shown in Fig. S4.† For the sample used at 300 °C, the typical MIL-100(Fe) phase was still the primary phase. In addition, the Fe₂O₃ phase was detected (2θ = 33.2 and 35.6° (PDF#33-0664)) for the two samples used at 325 and 350 °C. All of these results indicate that the decrease in NO_x conversion may have been due to the gradual decomposition of the MIL-100(Fe) structure above 300 °C. In addition, the N₂ selectivity was maintained at 100% over the temperature range of 100–260 °C. However, as the reaction temperature increased from 260 to 300 °C, the selectivity rapidly decreased, which generated increasing amounts of the unwanted N₂O by-product (Fig. S5a†). For V₂O₅–WO₃/TiO₂ catalyst, only 95% of the selectivity to N₂ was obtained when the high NO_x conversion occurred at 350 °C (Fig. S5b†). The concentration of N₂O over MIL-100(Fe) could not be detected in the active temperature range of 225–260 °C (NO_x conversion > 90%), For V₂O₅–WO₃/TiO₂ catalyst, it increased from 11 ppm to 63 ppm in the temperature range of 300–400 °C. These results indicated MIL-100(Fe) exhibited the higher selectivity to N₂ than the conventional V₂O₅–WO₃/TiO₂ catalyst at similar conversions in the SCR process.

3.2.2 Effect of SO₂ and H₂O. Because the existence of H₂O and SO₂ are unavoidable in the exhaust, it is important to investigate the effect of H₂O and SO₂ on the SCR activity of the catalyst. As shown in Fig. 3, the NO_x conversion is maintained at approximately 98% in the absence of H₂O and SO₂. When 500 ppm of SO₂ and 5% H₂O were introduced, the NO_x conversion decreased to 90% and a NO_x conversion of 88% over MIL-100(Fe) catalyst could be obtained after 10 h. It could be attributed to the competitive adsorption between SO₂ and reactants and the formed sulfate depositing on the surface of catalyst and then inhibiting the catalyst activity. After terminating the introduction of SO₂, the conversion of NO_x was restored to 93%, and high NO_x conversion greater than 91% was achieved after 6 h. After cutting off the supply of H₂O, the NO_x conversion raised to 95%. These results demonstrated MIL-100(Fe) catalyst exhibited a good capacity to resist SO₂ and H₂O, and the inhibition of SO₂ and H₂O on MIL-100(Fe) was a reversible process.

Fig. 3 The stability tests of SCR reaction at 250 °C over MIL-100(Fe). Reaction condition: 500 ppm NO, 500 ppm NH₃, 500 ppm SO₂ (when used), 5% H₂O (when used), 4% O₂ and N₂ balance, GHSV = 30 000 h⁻¹.

SEM images of the catalysts used in the durability test are shown in Fig. 1. The images indicate that the octahedral geometrical shape of the catalyst was retained after the durability test with and without SO₂ and H₂O (Fig. 1b and c). The XRD patterns of the catalysts used are shown in Fig. 4a. As the inset of Fig. 4a displayed, it was clearly demonstrated the MIL-100(Fe) catalyst were stable in the SCR reaction even with the presence of SO₂ and H₂O because all characteristic peaks of the used catalysts were still observed and the position was not changed compared with the as-synthesized MIL-100(Fe). Furthermore, the patterns of IR spectrum of MIL-100(Fe) catalysts before and after the durability tests were displaced in Fig. 4b. The characteristic peaks^24,33 (ν(C–O) at 1634 and 1378 cm⁻¹ and ν(C=C) at 1578 and 1453 cm⁻¹) of the used catalysts were similar with the catalyst MIL-100(Fe) before durability test, which was in accord with the results of XRD patterns of the as-synthesized and used MIL-100(Fe) catalysts. The positive peak (curve d) appeared at 1085 cm⁻¹ could be attributed to the deposited sulfate³⁴ on the surface of MIL-100(Fe) catalyst after durability test with SO₂ and H₂O.

Fig. 4 (a) Comparison of power X-ray diffraction patterns of various MIL-100(Fe) catalysts: (a) before activity test; (b) after durability test without SO₂ and H₂O; (c) after durability test with SO₂ and H₂O; (b) comparisons of IR spectrum of various MIL-100(Fe) catalyst: (a) before activity test; (b) after durability test without SO₂ and H₂O; (c) after durability test with SO₂ and H₂O; (d) difference spectra obtained by spectra a subtracting spectra c.; (c) XPS spectra of S 2p: (a) MIL-100(Fe) before durability test; (b) as-synthesized MIL-100(Fe) after durability test with SO₂ and H₂O.

The oxidation states of surface elements on MIL-100(Fe) catalysts before and after durability test were characterized using XPS. As shown in Fig. 4c, no S 2p band was observed on MIL-100(Fe) catalyst before durability test. After durability test with SO₂ and H₂O, an evident band at 186.6 eV attributed to S 2p³⁵ was observed and the S content was calculated to be 1.15% (shown in Table 2), implying the formation of sulfur-containing species on the catalyst surface. Moreover, the Fe content in the used catalyst declined compared with the fresh MIL-100(Fe) catalyst. The reason could be attributed to the coverage of the formed sulfate on the iron sites. Furthermore, the C/O ratio of the used MIL-100(Fe) catalyst was similar with that of the fresh MIL-100(Fe). The result indicated the effect of organic coordination (i.e. the residual trimesic acid) in MIL-100(Fe) on the SCR catalytic activity could be ignored.

Table 2 Ratio of Fe, C, O, F and S elements on MIL-100(Fe) before and after durability test (wt%)^a

Catalysts	Fe	C	O	F	S
a a, MIL-100(Fe) before durability test; b, MIL-100(Fe) after durability test with SO2 and H2O.
a	4.59	65.85	29.01	0.55	—
b	3.47	65.98	29.26	0.14	1.15

---

To elucidate the surface acidity on fresh and used MIL-100(Fe), in situ DRIFT spectra of NH₃ adsorption were recorded. Prior to the test, the catalysts (0.01 g) were exposed in NH₃/N₂ (500 ppm) for 1 h after treating with N₂ for 30 min to evacuate the impurity at 250 °C, and followed by purging with N₂ for 30 min at 250 °C and the results were displayed in Fig. S6.† The bands appeared at 1735, 1662 and 1455 cm⁻¹ were attributed to asymmetric and symmetric bending vibrations of NH₄⁺ species on Brønsted acid sites.^36,37 The bands displayed at 1603, 1296 and 1151 cm⁻¹ were assigned to coordinated NH₃ on Lewis acid sites.³⁸ It is well known the relative quantity of Brønsted acid sites (1735 and 1662 cm⁻¹ assigned to δ_s(NH₄⁺)) and Lewis acid sites (1296 and 1151 cm⁻¹ assigned to δ_s(NH₃)) on surface of MIL-100(Fe) catalyst could be calculated by the corresponding integral characteristic peak area. According to Fig. S6b,† the results indicated both the intensity of B and L acid sites of the used MIL-100(Fe) decreased compared with the fresh MIL-100(Fe). The deposited sulfate on the active sites could prevent the adsorption and activation of gaseous NH₃, which induced the decrease of quantity of acid sites on MIL-100(Fe).

3.3 In situ DRIFT studies

To investigate the catalytic reaction mechanism, DRIFTS experiments were performed to identify the species absorbed on the catalyst surface and deduce the surface reaction under actual reaction conditions.

3.3.1 IR spectra of NH₃ adsorption on MIL-100(Fe). Prior to the DRIFT tests, MIL-100(Fe) was first treated at 250 °C for 10 h in N₂ and then cooled to 100 °C for adsorption of NH₃/N₂. Fig. 5 shows the DRIFT spectra of MIL-100(Fe) after it was exposed to 500 ppm NH₃/N₂ from 100 to 275 °C. The bands at 1735, 1662 and 1455 cm⁻¹ were attributed to symmetric and asymmetric bending vibrations of NH₄⁺ species on Brønsted acid sites,^36,37 and the intensity of these bands increased with temperature above 200 °C. The bands at 1603, 1296 and 1151 cm⁻¹ were assigned to coordinated NH₃ on Lewis acid sites.³⁸ The observed results were consistent with these as demonstrate in Fig. S6.† The bands at 3353, 3250 and 3150 cm⁻¹ corresponding to N–H stretching were observed.³⁹ The structure of MIL-100(Fe) is based on μ₃-oxo-centered trimers of octahedral Fe(III). In addition, two of the three octahedral iron atoms are linked by H₂O molecules, and the third iron atom is coordinated to either F⁻ or OH⁻ atoms.²² The anions (OH⁻) connected to the iron sites can act as Brønsted acid sites. Due to the treatment at 250 °C in N₂ for 10 h, some of the coordinated water molecules were removed, and a large number of Fe(III) CUS were formed, which could act as Lewis acid sites.^23,30 In addition, the band at 1506 cm⁻¹ was due to the weak adsorption of amide species (NH₂),⁴⁰ and the bands at 1551 and 1375 cm⁻¹ were assigned to an intermediate formed during the oxidation of ammonia.^41–43

Fig. 5 DRIFT spectra of MIL-100(Fe) exposed to NH₃ at various temperatures. Reaction condition: [NH₃] = 500 ppm, and N₂ balance.

3.3.2 IR spectra of NO and O₂ adsorption on MIL-100(Fe). NO adsorption on MIL-100(Fe) was investigated by DRIFT spectroscopy. Fig. 6 shows the spectra obtained after the introduction of 500 ppm + 4% O₂ at different temperatures. The bands detected at 1605, 1550, 1278 and 1212 cm⁻¹ were attributed to the nitrate species.^44,45 The band at 1901 cm⁻¹ was attributed to the nitrosyl group, whose formation could be due to the coordination of a NO molecule to a Lewis acid site of Fe³⁺ via the nitrogen atom.⁴⁶ It should be noted that the intensity of the nitrosyl band decreased with increasing temperature and disappeared at 250 °C. Simultaneously, two new bands at 1624 and 1842 cm⁻¹ appeared. The band at 1624 cm⁻¹ was due to NO₂ species.⁴⁷ According to the results reported in the literature,²³ the activation of MIL-100(Fe) at 250 °C could form Fe(II) CUS due to the departure of anionic ligands (F⁻ and OH⁻). In addition, the band at 1842 cm⁻¹ corresponds to the nitrosyl absorbed on Fe²⁺ sites.⁴⁶ The results suggested that Fe(III) CUS could be reduced to Fe(II) CUS above 250 °C during NO adsorption. In combination with the formation of NO₂, the transformation of the adsorptive species was as follows:

Fig. 6 DRIFT spectra of MIL-100(Fe) exposed to NO + O₂ at various temperatures. Reaction condition: [NO] = 500 ppm, [O₂] = 4% and N₂ balance.

3.3.3 IR spectra of the reaction species between the adsorbed NH₃ species and NO + O₂. To understand the SCR reaction mechanism on the surface of MIL-100(Fe), a DRIFT study of the reaction between NO + O₂ and the pre-adsorbed NH₃ species was performed at 250 °C. In this experiment, the catalyst was first treated with 500 ppm NH₃/N₂ at 250 °C for 1 h, followed by N₂ purging for 0.5 h, and then 500 ppm NO + 4% O₂/N₂ was passed over the sample. The DRIFT spectra were recorded as a function of time and are shown in Fig. 7. After the introduction of NO + O₂, both bands corresponding to NH₄⁺ species and coordinated NH₃ decreased in intensity. This result indicated that both ammonia species, which act as reducing agents, reduce NO_x in the SCR reaction. In addition, the bands assigned to the NH₄⁺ species (1735 and 1455 cm⁻¹) disappeared after 4 min, implying that NH₄⁺ was prior to react with NO_x compared to the coordinated NH₃. Furthermore, the amide species (NH₂) may be an important intermediate in the SCR reaction because the amide bands immediately disappeared when NO + O₂ was passed over MIL-100(Fe). The band at 3658 cm⁻¹, which was assigned to H₂O, appeared, and the intensity of this band gradually increased over time. After the active NH₃ species were completely consumed, the nitrate species (1605, 1550 and 1278 cm⁻¹) began to accumulate on the catalyst surface.

Fig. 7 DRIFT spectra of MIL-100(Fe) exposed to NO_x after NH₃ adsorption and purged with N₂ at 250 °C for various times. Reaction condition: [NH₃] = 500 ppm, [NO] = 500 ppm, [O₂] = 4% and N₂ balance.

3.3.4 IR spectra of the reaction species between the adsorbed NO_x species and NH₃. A DRIFT study of the reaction between NH₃ and the pre-adsorbed NO_x species on MIL-100(Fe) at 250 °C was performed. In this test, the MIL-100(Fe) catalyst was treated with 500 ppm NO + 4% O₂/N₂ at 250 °C for 1 h, followed by purging with N₂ for 0.5 h, and then, 500 ppm NH₃ was introduced. The results are shown in Fig. 8. The introduction of NH₃ resulted in the rapid disappearance of NO₂ species, which implied that this species was reactive in the SCR reaction. In contrast, the nitrate species (1212 cm⁻¹) were always present on the surface even after 30 min, suggesting that they could have been inactive. Simultaneously, the bands assigned to the coordinated NH₃ (1603, 1296 and 1151 cm⁻¹) and NH₄⁺ (1735 and 1662 cm⁻¹) species were observed.

Fig. 8 DRIFT spectra of MIL-100(Fe) exposed to NH₃ after NO and O₂ co-adsorption and purged with N₂ at 250 °C for various times. Reaction condition: [NH₃] = 500 ppm, [NO] = 500 ppm, [O₂] = 4% and N₂ balance.

3.3.5 IR spectra of the SCR reaction. To investigate the surface reactive species under realistic SCR reaction conditions, the DRIFT spectra were recorded in a flow of 500 ppm NO + 4% O₂ + 500 ppm NH₃ over the MIL-100(Fe) catalyst at 250 °C. As shown in Fig. 9, the bands attributed to the NH₄⁺ species (1735, 1662 and 1455 cm⁻¹) and coordinated NH₃ (1603, 1296 and 1188 cm⁻¹) were observed as the reaction time increased. The bands assigned to the NO_x species were not observed during the reaction. The characteristic peaks attributed to H₂O generated from the SCR reaction (3658 cm⁻¹) were detected.

Fig. 9 DRIFT spectra of MIL-100(Fe) exposed to the SCR reaction at 250 °C for various times. Reaction condition: [NH₃] = 500 ppm, [NO_x] = 500 ppm, [O₂] = 4% and N₂ balance.

The DRIFT results for NH₃ adsorption indicated that both Brønsted and Lewis acid sites existed on the surface of MIL-100(Fe), and NH₃ was primarily adsorbed on MIL-100(Fe) in the form of ionic NH₄⁺ and coordinated NH₃. In addition, after the introduction of NO + O₂, the disappearance of the NH₄⁺ species indicated that ionic NH₄⁺ may have been the primary reactive ammonia species adsorbed on MIL-100(Fe) during the reaction.

Because NO was easily oxidized by O₂ on Fe³⁺,⁴⁷ NO was adsorbed on the iron sites to form NO₂ species (as described in Section 3.3.2). Then, the reactive NO₂ could react with the neighboring NH₄⁺ on the iron sites to form N₂. Therefore, we propose that a Langmuir–Hinshelwood (L–H) mechanism was dominant in the SCR reaction on MIL-100(Fe), as described by the following reaction scheme.

NH₃(g) → NH₃(a) → NH⁺₄(a) → NH⁺₄(a) + NO₂(a) → N₂(g) + H₂O(g) (5)

During this process, Fe(III) CUS could be reduced to Fe(II) CUS in the oxidation step of adsorbed NO to NO₂, and then, Fe(II) was reoxidized to Fe(III) by O₂ to complete the redox cycle.

The coordinated NH₃ was considered to be less reactive due to the high activation energy required to form NH₂ species.⁴⁸ However, based on the results presented in Section 3.3.1, the coordinated NH₃ adsorbed on MIL-100(Fe) could undergo oxidative dehydrogenation to form NH₂ species. In addition, the NH₂ species could react with NO (shown in Fig. 5), suggesting that the SCR reaction on MIL-100(Fe) could also occur via the Eley–Rideal (E–R) mechanism.

NH₃(g) → NH₃(a) → NH₂(a) → NH₂(a) + NO(g) → N₂(g) + H₂O(g) (7)

The stable structure of the MIL-100(Fe) catalyst used in the durability test indicated that SO₂ did not easily react with the iron active sites. In addition, the absence of sulfate on the surface of used MIL-100(Fe) catalyst after durability test suggested that the porous structure of MIL-100(Fe) may prevent the deposition of sulfate. Therefore, a high conversion of NO_x could be recovered by eliminating SO₂. The declined NO_x conversion may be attributed to the competitive adsorption between SO₂ and reactants.

4. Conclusions

The porous MOF MIL-100(Fe) catalyst was observed to be an excellent catalyst and exhibited better performance in the SCR of NO_x with NH₃ at lower temperatures (245–300 °C) than those required by the conventional V₂O₅–WO₃/TiO₂ catalyst. In addition, the MIL-100(Fe) catalyst exhibited satisfactory stability in the presence of H₂O and SO₂ and good self-regeneration capability for the SCR of NO_x after eliminating H₂O and SO₂. XRD, IR and SEM analyses indicated that the structure of the catalyst used in the durability test was the same as that of the fresh sample, suggesting that MIL-100(Fe) was stable during the SCR process below 300 °C. Based on the DRIFT results, two reaction pathways were proposed. The reaction between the NH₄⁺ species and the adsorbed NO₂ species on Fe(II) CUS was the primary reaction pathway, which is consistent with the L–H mechanism. The SCR process also followed the E–R mechanism, through which the amide species reacted with gaseous NO. The valance change of Fe promoted the formation of adsorbed NO₂ species, which were beneficial for the SCR reaction. This study provides insight that may benefit future investigations of MOFs as high-efficiency catalysts in the SCR process. In addition, this work supports the further study of high NO_x conversion over MOFs at low temperature.

Acknowledgements

This work was supported by the National Basic Research Program of China (2011CB936002) and the National Natural Science Foundation of China (no. 21277015).

Notes and references

1. L. J. Alemany, F. Berti, G. Busca, G. Ramis, D. Robba, G. P. Toledo and M. Trombetta, Appl. Catal., B, 1996, 10, 299–311.
2. L. Lietti, I. Nova and P. Forzatti, Top. Catal., 2000, 11/12, 111–122.
3. L. Lietti, Appl. Catal., B, 1996, 10, 281–297.
4. L. Chen, J. Li and M. Ge, J. Phys. Chem. C, 2009, 113, 21177–21184.
5. W. Shan, F. Liu, H. He, X. Shi and C. Zhang, Chem. Commun., 2011, 47, 8046–8048.
6. J. Xue, X. Wang, G. Qi, J. Wang, M. Shen and W. Li, J. Catal., 2013, 297, 56–64.
7. P. R. Ettireddy, N. Ettireddy, T. Boningari, R. Pardemann and P. G. Smirniotis, J. Catal., 2012, 292, 53–63.
8. Y. J. Kim, H. J. Kwon, I. Heo, I. Nam, B. K. Cho, J. W. Choung, M. Cha and G. K. Yeo, Appl. Catal., B, 2012, 126, 9–21.
9. J. Park, H. J. Park, J. H. Baik, I. Nam, C. Shin, J. Lee, B. K. Cho and S. H. Oh, J. Catal., 2006, 240, 47–57.
10. T. Boningari, R. Koirala and P. G. Smirniotis, Appl. Catal., B, 2013, 140–141, 289–298.
11. W. S. Kijlstra, M. Biervliet, E. K. Poels and A. Bliek, Appl. Catal., B, 1998, 16, 327–337.
12. J. Y. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450–1459.
13. N. Stock and S. Biswas, Chem. Rev., 2012, 112, 933–969.
14. L. S. Cheng, R. T. Yang and N. Chen, J. Catal., 1996, 164, 70–81.
15. L. Y. Kong, Z. H. Zhang, H. F. Zhu, H. Kawaguchi, T. Okamura, M. Doi, Q. Chu, W. Y. Sun and N. Ueyama, Angew. Chem., Int. Ed., 2005, 44, 4352–4355.
16. X. S. Wang, S. Q. Ma, D. F. Sun, S. Parkin and H. C. Zhou, J. Am. Chem. Soc., 2006, 128, 16474–16475.
17. R. Q. Zou, H. Sakurai, S. Han, R. Q. Zhong and Q. Xu, J. Am. Chem. Soc., 2007, 129, 8402–8403.
18. S. Han, Y. Huang, T. Watanabe, S. Nair, K. S. Walton, D. S. Sholl and J. C. Meredith, Microporous Mesoporous Mater., 2013, 173, 86–91.
19. N. Apostolescu, B. Geiger, K. Hizbullah, M. T. Jan, S. Kureti, D. Reichert, F. Schott and W. Weisweiler, Appl. Catal., B, 2006, 62, 104–114.
20. M. Iwasaki, K. Yamazaki, K. Banno and H. Shinjoh, J. Catal., 2008, 260, 205–216.
21. F. Liu, H. He and C. Zhang, Chem. Commun., 2008, 2043–2045.
22. P. Horcajada, S. Surblé, C. Serre, D. Y. Hong, Y. K. Seo, J. S. Chang, J. M. Grenèche, I. Margiolaki and G. Férey, Chem. Commun., 2007, 2820–2822.
23. J. W. Yoon, Y. K. Seo, Y. K. Hwang, J. S. Chang, H. Leclerc, S. Wuttke, P. Bazin, A. Vimont, M. Daturi, E. Bloch, P. L. Llewellyn, C. Serre, P. Horcajada, J. M. Grenèche, A. E. Rodrigues and G. Férey, Angew. Chem., Int. Ed., 2010, 49, 5949–5952.
24. S. Wuttke, P. Bazin, A. Vimont, C. Serre, Y. Seo, Y. K. Hwang, J. Chang, G. Férey and M. Daturi, Chem.–Eur. J., 2012, 18, 11959–11967.
25. C. Petit and T. J. Bandosz, Adv. Funct. Mater., 2011, 21, 2108–2117.
26. E. Lenoir, C. Vagner, J. W. Yoon, P. Bazin, F. Ragon, Y. K. Hwang, C. Serre, J. Chang and P. L. Llewellyn, J. Am. Chem. Soc., 2012, 134, 10174–10181.
27. P. Sazama, B. Wichterlová, E. Tábor, P. Šťastný, N. K. Sathu, Z. Sobalík, J. Dědček, Š. Sklenák, P. Klein and A. Vondrová, J. Catal., 2014, 312, 123–138.
28. B. H. Toby, J. Appl. Crystallogr., 2001, 34, 210–213.
29. X. Fan, F. Qiu, H. S. Yang, W. Tian, T. F. Hou and X. B. Zhang, Catal. Commun., 2011, 12, 1298–1301.
30. R. Canioni, C. R. Marchal, F. Sécheresse, P. Horcajada, C. Serre, M. Hardi-Dan, G. Férey, J. M. Grenèche, F. Lefebvre, J. S. Chang, Y. K. Hwang, O. Lebedev, S. Turner and G. V. Tendeloo, J. Mater. Chem., 2011, 21, 1226–1233.
31. S. Yang, C. Wang, L. Ma, Y. Peng, Z. Qu, N. Yan, J. Chen, H. Chang and J. Li, Catal. Sci. Technol., 2013, 3, 161–168.
32. P. Balle, B. Geiger and S. Kureti, Appl. Catal., B, 2009, 85, 109–119.
33. M. Haouas, C. Volkringer, T. Loiseau, G. Férey and F. Taulelle, J. Phys. Chem. C, 2011, 115, 17934–17944.
34. T. Yamaguchi, T. Jin and K. Tanabe, J. Phys. Chem., 1986, 90, 3148–3152.
35. M. S. Maqbool, A. K. Pullur and H. P. Ha, Appl. Catal., B, 2014, 152–153, 28–37.
36. S. D. Lin, A. C. Gluhoi and B. E. Nieuwenhuys, Catal. Today, 2004, 90, 3–14.
37. B. Jiang, Z. Li and S. Lee, Chem.–Eur. J., 2013, 225, 52–58.
38. G. Ramis, L. Yi and G. Busca, Catal. Today, 1996, 28, 373–380.
39. J. Y. Luo, H. Oh, C. Henry and W. Epling, Appl. Catal., B, 2012, 123–124, 296–305.
40. W. S. Kijlstra, D. S. Brands, E. K. Poels and A. Bliek, J. Catal., 1997, 171, 219–230.
41. G. Ramis, L. Yi, G. Busca, M. Turco, E. Kotur and R. J. Willey, J. Catal., 1995, 157, 523–535.
42. M. A. Larrubia, G. Ramis and G. Busca, Appl. Catal., B, 2001, 30, 101–110.
43. M. A. Larrubia, G. Ramis and G. Busca, Appl. Catal., B, 2000, 27, L145–L151.
44. W. S. Kijlstra, D. S. Brands, E. K. Poels and A. Bliek, J. Catal., 1997, 171, 208–218.
45. M. Machida, M. Uto, D. Kurogi and T. Kijima, Chem. Mater., 2000, 12, 3158–3164.
46. S. Wuttke, P. Bazin, A. Vimont, C. Serre, Y. Seo, Y. K. Hwang, J. Chang, G. Férey and M. Daturi, Chem.–Eur. J., 2012, 18, 11959–11967.
47. R. Q. Long and R. T. Yang, J. Catal., 2002, 207, 224–231.
48. G. Qi and R. T. Yang, J. Phys. Chem. B, 2004, 108, 15738–15747.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra07028c

---