Effect of initial support particle size of MnO_x/TiO₂ catalysts in the selective catalytic reduction of NO with NH₃

Abstract

A series of manganese-based catalysts supported by 5–10 nm, 10–25 nm, 40 nm and 60 nm anatase TiO₂ particles was synthesized via an impregnation method to investigate the effect of the initial support particle size on the selective catalytic reduction (SCR) of NO with NH₃. All catalysts were characterized by transmission electron microscopy (TEM), N₂ physisorption/desorption, X-ray diffraction (XRD), temperature programmed techniques, X-ray photoelectron spectroscopy (XPS) and in situ diffuse reflectance infrared transform spectroscopy (DRIFTS). TEM results indicated that the particle sizes of the MnO_x/TiO₂ catalysts were similar after the calcination process, although the initial TiO₂ support particle sizes were different. However, the initial TiO₂ support particle sizes were found to have a significant influence on the SCR catalytic performance. XPS and NH₃-TPD results of the MnO_x/TiO₂ catalysts illustrated that the surface Mn⁴⁺/Mn molar ratio and acid amount could be influenced by the initial TiO₂ support particle sizes. The order of surface Mn⁴⁺/Mn molar ratio and acid amount over the MnO_x/TiO₂ catalysts was as follows: MnO_x/TiO₂(10–25) > MnO_x/TiO₂(40) > MnO_x/TiO₂(60) > MnO_x/TiO₂(5–10), which agreed well with the order of SCR performance. In situ DRIFTS results revealed that the NH₃-SCR reactions over MnO_x/TiO₂ at low temperature occurred via a Langmuir–Hinshelwood mechanism. More importantly, it was found that the bridge and bidentate nitrates were the main active substances for the low-temperature SCR reaction, and bridge nitrate adsorbed on Mn⁴⁺ showed superior SCR activity among all the adsorbed NO_x species. The variation of the initial TiO₂ support particle size over MnO_x/TiO₂ could change the surface Mn⁴⁺/Mn molar ratio, which could influence the adsorption of NO_x species, thus bringing about the diversity of the SCR catalytic performance.

---

1 Introduction

Nitrogen oxides (NO_x), as a harmful exhaust emission, can induce various human health and environmental hazards, such as acid rain, photochemical smog, and ozone layer depletion.^1,2 Increasing attention has been focused on NO_x abatement, including NO_x storage-reduction (NSR or LNT),^3–5 NO_x decomposition, and Selective Catalytic Reduction (SCR).^6–8 SCR technology has been regarded as the most promising way to remove NO_x, and has been commercially applied in diesel engines and power plants. However, due to increasingly strict emission legislation, it is crucial to develop future catalysts for low-temperature NO_x abatement techniques, to meet future exhaust gas emission legislation.⁹

Because of their various valence states and low cost, transition metal oxides (MnO_x, FeO_x, CoO_x, CuO_x), are usually used as catalysts in SCR research.^6–18 Owing to their inherent instability and poor anti-sintering ability, transition metal oxides are usually supported on an “inert” support to enhance their thermal stability and catalyst lifespan. Pena et al.¹⁹ investigated low-temperature SCR using VO_x, CrO_x, MnO_x, FeO_x, CoO_x, NiO_x and CuO_x catalysts supported on TiO₂. The SCR activity for the various transition metal oxides as active sites supported on TiO_x decreased in the following order: MnO_x > CuO_x ≥ CrO_x ≥ CoO_x > FeO_x ≫ VO_x ⋙ NiO_x, indicating that MnO_x was the most promising catalyst for future practical applications. Among supported catalysts, the properties of the support and the interaction between MnO_x and the support are often considered to be the main parameters affecting the catalytic activity. Smirniotis et al.²⁰ synthesized a series of manganese oxides supported on TiO₂, Al₂O₃, and SiO₂ to investigate the influence of the support on SCR activity with a space velocity 8000 h⁻¹. They found that the SCR activity of the supported MnO_x catalysts decreased in the following order: TiO₂ (anatase) > SiO₂ > TiO₂ (rutile) > TiO₂ (anatase, rutile) > γ-Al₂O₃. MnO_x/TiO₂ (anatase) showed excellent NO_x abatement activity, even at 120 °C when it reached complete conversion. Therefore, MnO_x/TiO₂ catalysts have been investigated on the basis of the effects of the preparation method,²¹ manganese precursors²² and manganese loading²³ on SCR activity.^19,24 However, the interaction between manganese oxides and TiO₂ has not been fully investigated, especially on the effect of the support size.

The interaction between the active component and the support is not only influenced by the support type, but also relies on the size of the support.^25–30 Soykal et al.³⁰ studied the effect of support particle size in steam reforming of ethanol over CoO_x/CeO₂ catalysts with CeO₂ support particle sizes in the micro- and nano-range. CoO_x/CeO_x with smaller CeO₂ particle sizes manifested superior catalytic performance and coking resistance ability. These results can be ascribed to a combination of factors, including reducibility, dispersion and redox sites on the catalyst surface. All of these properties could be affected by the particle size of the support, thus leading to different catalytic performances. Pakulska et al.²⁸ investigated the effect of support particle size over NiO/ZrO₂ and NiO/CeO₂ on the catalytic oxidation of methane. They found that zirconia particles size could change the number of catalytic sites via varying the pre-exponential factor and ceria size could change the reaction activation energy by varying oxygen transport to the active sites. Xu et al.^25–27 studied a series of Ni/ZrO₂ catalysts with various ZrO₂ particle sizes to systematically investigate the effect of support particle size on the reforming of methane with CO₂. Their proposed nanocomposite catalyst, which could be defined as a catalyst with comparable Ni metal and zirconia nanocrystals, showed superior catalytic activity and long-term stability. They attributed this excellent catalytic performance to the boundary or perimeter between Ni and ZrO₂, which could easily be generated by similar-sized Ni and ZrO₂ particles. All of the above examples illustrate that the support size can significantly influence the catalytic performance and catalyst lifespan, by affecting the properties of active sites or the boundary between active sites and the support. However, up until now, corresponding research regarding the effect of support size on NO_x abatement has rarely been reported.

In this work, a series of MnO_x/TiO₂ catalysts was prepared via an impregnation method with a variation of TiO₂ support particle sizes to investigate the effect of the initial TiO₂ support size on the SCR performance. In addition, these catalysts were characterized by N₂ physisorption, TEM, XRD, H₂-TPR, O₂-TPD, XPS and DRIFTS, to understand the effects of the structure, redox properties and active site-support interaction on the SCR catalysis.

2 Experimental

2.1 Preparation of catalysts

Anatase TiO₂ particles of various sizes (Aladdin Reagent Co., Ltd.) were used as support materials without any purification. All of the catalysts were synthesized via a wet impregnation method. In a typical process, 2 g TiO₂ with various particle sizes and 0.84 mL 50 wt% Mn(NO₃)₂ (Tianjin Fuchen chemical reagents company) were added into 20 mL deionized water, and the mixture was vigorously stirred for 24 h at room temperature. Then the solution was dried under vacuum at 60 °C in a rotary evaporator. The obtained powder was dried at 110 °C overnight, and then calcined at 550 °C for 5 h with a heating rate of 1 °C min⁻¹ in a muffle furnace. The obtained materials were denoted as MnO_x/TiO₂(n), where n represents the particle size of TiO₂. The actual manganese loading of the catalysts was determined by X-ray fluorescence spectrometry (XRF).

2.2 Catalytic test

A catalytic performance test for the NH₃-SCR reaction was implemented in a fixed-bed plug-flow stainless-steel tubular reactor (L = 60 cm, Φ_in = 7 mm). The typical components of the simulated gas were as follows: 500 ppm NO, 500 ppm NH₃, 5 vol% O₂ and balance Ar with a space velocity of 40 000 mL g⁻¹ h⁻¹. The reaction temperature was recorded by a K-type thermocouple located inside the tube near the catalyst bed. Prior to the SCR catalytic measurements, the catalyst was pretreated in 20 vol% O₂/Ar at 400 °C for 1 h. The reaction was carried out from 50 °C to 350 °C with an increment of 25 °C and each temperature was held for 1 h to obtain a stable reaction conversion. The concentration of NO_x was detected by a fuel analyzer (AFRISO, M 60). NO_x conversion was calculated using the following equation:

2.3 Catalyst characterization

N₂ adsorption–desorption isotherms were collected at −196 °C on a BELSORP-Max analyzer. Prior to each measurement, the catalyst was degassed under high vacuum at 300 °C for 3 h. The surface area of each sample was obtained using the Brunauer–Emmett–Teller (BET) method and the pore size distribution was determined via the Barrett–Joyner–Halenda (BJH) method using desorption branch.

X-ray diffraction (XRD) patterns were measured on a SHIMADZU XRD-6100 Powder diffractometer using a Cu Kα radiation source (λ = 0.15406 nm). The diffraction patterns were recorded over a 2θ range of 10–80° with a scan speed of 6 ° min⁻¹ and a step size of 0.02°.

H₂-temperature programmed reduction (H₂-TPR) experiments were performed on a Quantachrome ChemBET Pulsar chemisorption analyzer with a thermal conductivity detector (TCD). H₂-TPR curves were acquired in 5 vol% H₂/He from 50 °C to 800 °C with a temperature rate of 10 °C min⁻¹ and held at 800 °C for 30 min. Prior to each measurement, about 0.040 g sample was pretreated from room temperature to 500 °C with a temperature rate of 10 °C min⁻¹ in a flowing purified He gas (30 mL min⁻¹) and was then kept at 500 °C for 30 min, in order to remove weakly adsorbed moisture and gas on the catalyst surface. After the sample was cooled to 50 °C, the flowing gas was switched to 5 vol% H₂ + He (30 mL min⁻¹) and the sample was purged for 30 min until the TCD signal was stable. NH₃-temperature programmed desorption (NH₃-TPD) experiments were carried out using the same equipment. NH₃-TPD curves were obtained in flowing He (30 mL min⁻¹) from 50 °C to 800 °C with a temperature ramp rate of 10 °C min⁻¹, and 800 °C was held for 30 min. The pretreatment procedure for NH₃-TPD was similar to that of H₂-TPR. Adsorption of NH₃ was conducted under flowing 5 vol% NH₃ + He for ca. 30 min at 100 °C, and then purged with flowing He until the baseline became stable.

O₂-TPD experiments were performed on a mass spectrometer (TILON GRP TECHNOLOGY LIMITED LC-D200M). O₂-TPD curves were obtained under flowing N₂ (30 mL min⁻¹) from 100 °C to 800 °C with a temperature ramp rate of 10 °C min⁻¹, and 800 °C was held for 30 min. Prior to each measurement, about 100 mg of sample was pretreated in a U-shaped quartz tube at 500 °C for 30 min under flowing N₂ (30 mL min⁻¹), followed by cooling to 100 °C. Adsorption of oxygen was then conducted at 20 vol% O₂ + N₂ for ca. 30 min at 100 °C until the O₂ signal became stable. After purging with flowing N₂ until the baseline signal reached a stable state, the O₂-TPD signals began to be recorded with the mass-to-charge ratios (m/z = 16 and 32).

Transmission Electron Microcopy (TEM) was conducted on a Tecnai G2 F20 transmission electron microscope. More than 200 particles were randomly measured to determine the mean diameter of TiO₂ and MnO_x/TiO₂ catalysts, using the equation d = ∑(n_id_i)/∑n_i, where n_i is the number of particles with a diameter of d_i.

The components of the catalysts were examined on a Bruker S8 TIGERX X-Ray Fluorescence analyzer (XRF), equipped with a 60 kV X-ray tube (Rh Kα radiation) with a maximum power of 4 W, and a maximum current of 170 mA. The XPS spectra were conducted on a Kratos Analytical AXIS Ultra DLD spectrometer with an Al Kα radiation source (hν = 1253.6 eV), operated at an accelerating power of 15 kW. The samples were degassed with vacuum-pumping to 5 × 10⁻⁷ Pa before the measurement. The binding energy was calibrated by the C 1s peak at 284.8 eV and the final spectra were obtained by subtracting a Shirley-type background.

In situ diffusion reflection infrared Fourier transformed spectroscopy (DRIFTS) experiments were conducted on a Nicolet iS50 FTIR spectrometer equipped with a DRIFT cell containing ZnSe windows and MCT detector.³¹ The spectra were obtained from 4000 to 525 cm⁻¹ at a resolution of 2 cm⁻¹ by accumulating 32 scans. The temperature was controlled by a Pike temperature controller and the gas composition was governed by mass flow control (MFC). The catalyst was pretreated at 500 °C for 1 h under flowing 20 vol% O₂ + N₂ (40 mL min⁻¹) and was then cooled down to 100 °C. The background spectrum was recorded under flowing N₂ at 100 °C, which was further subtracted from the sample spectrum.

3 Results and discussion

3.1 Catalytic performance

Fig. 1 shows the NH₃-SCR catalytic performance of the MnO_x/TiO₂(n) catalysts with various TiO₂ particle sizes. For all the catalysts, NO_x conversion firstly increased gradually from 65 °C to 265–290 °C, and then decreased steadily with further increasing of the temperature to 400 °C. The maximum NO_x conversion temperature was 265 °C for MnO_x/TiO₂(5–10), MnO_x/TiO₂(10–25) and MnO_x/TiO₂(40) and 290 °C for MnO_x/TiO₂(60). These results can be attributed to various SCR reaction pathways based on the NO₂/NO ratio.^6,7,9,10,32 The pathway of the NH₃-SCR reaction was regarded to occur via three reaction routes; namely, standard-, fast- and slow-SCR, which rest on the NO₂/NO fraction in the reaction feed.⁷

4NH₃ + 4NO + O₂ → 4N₂ + 6H₂O, standard SCR

4NH₃ + 2NO + 2NO₂ → 2N₂ + 6H₂O, fast SCR

4NH₃ + 2NO₂ + O₂ → ON₂ + 6H₂O, slow SCR

Fig. 1 NO_x conversion (A), T50 and T80 (B) of MnO_x/TiO₂(n) with various TiO₂ particle sizes.

As shown in Fig. 1, at the initial stage, the reaction occurred through a standard route. With increasing reaction temperature, the NO₂/NO ratio increased through NO oxidation, inducing a variation in the main reaction pathway from the standard route to fast SCR. With increasing the reaction temperature, the ratio of NO₂/NO increased above 1, causing the main reaction pathway to change to slow SCR from fast SCR.^33,34 By further increasing reaction temperature, the NO₂/NO ratio decreased with the restriction of the thermal equilibrium of NO₂ decomposition. The main reaction pathway changed to the standard SCR, thus decreasing the SCR catalytic activity.³³

In order to compare the catalytic performance of the various catalysts, T50 and T80 (the temperatures of NO_x conversion at 50% and 80%) were introduced and are shown in Fig. 1B. T50 decreased from 160 °C to 155 °C and 135 °C, when the TiO₂ particle sizes decreased from 60 nm to 40 nm and 10–25 nm. Nevertheless, by further decreasing the TiO₂ particle size from 10–25 nm to 5–10 nm, T50 increased from 135 °C to 185 °C. A similar trend could be found in T80 as T50 over the corresponding catalysts. The catalytic activity of the MnO_x/TiO₂(n) catalysts showed a volcano-type dependence on the TiO₂ particle size, reaching a maximum at 10–25 nm. These results illustrated that the SCR performance of the MnO_x/TiO₂(n) catalysts varied with the TiO₂ support particle size. Soykal et al.³⁰ reported that the support particle size could change the reducibility and relative amount of redox sites on the catalyst surface. Generally, the SCR activity could be dependent on the surface acidity property, redox property and their synergistic effects.^6,32 The various initial TiO₂ support sizes could affect the characteristics of the manganese oxide active sites, and could thereby have an impact on the SCR catalytic performance. Therefore, the characteristics of the supported manganese catalysts, such as redox, acidity and their synergistic effects, were then investigated to assess the effect of initial TiO₂ support particle size on the SCR performance of the MnO_x/TiO₂(n) catalysts.

3.2 TEM

Fig. 2 shows TEM images of various TiO₂ support sizes. The size distribution of the various TiO₂ supports was 8.2 ± 3.5 nm, 18.3 ± 7.4 nm, 43.5 ± 5.2 nm and 63.6 ± 5.2 nm, similar to as purchased. However, the size distributions of the various MnO_x/TiO₂ catalysts, as shown in Fig. S1,† were similar. All the MnO_x/TiO₂ catalysts had an average diameter of ca. 55 nm. These results revealed that the impregnation and calcination process resulted in the variation in the particle sizes of the MnO_x/TiO₂ catalysts.

Fig. 2 TEM images of the various TiO₂(n) supports of n = 5–10 (A), 10–25 (B), 40 (C) and 60 (D).

3.3 XRD

Fig. 3 shows the XRD patterns of MnO_x/TiO₂(n) with various TiO₂ particle sizes. All of the catalysts showed the same 2θ diffraction positions and different diffraction intensities. The diffraction of 2θ at 25.3°, 37.8°, 48.0°, 53.9°, 55.1°, 62.8°, 68.9°, 70.5° and 75.2° could be assigned to (101), (004), (200), (105), (211), (204), (116), (220) and (215) planes of anatase-TiO₂ (JCPDS, 65-5714), respectively. The theoretical monolayer coverage amount could be estimated from structural calculations.^35–37 Based on the cation density of MnO_x, a reference value of 15 Ų per MnO_x unit was used to calculate the monolayer surface coverage.^35–37 The virtual loading was ca. 3.1 wt% Mn for monolayer surface coverage, according to the surface area of our investigated samples. Mn loading in our investigated catalysts was 10 wt%, which was more than three times that of the monolayer manganese amount. However, no independent diffraction peaks corresponding to manganese oxides could be observed in all the samples, indicating that the incorporated manganese oxides were in a highly dispersed state or in an amorphous or poorly crystalline state. These results could be ascribed to the strong interaction between Mn and anatase TiO₂, which is in agreement with the previous reports.^20,23

Fig. 3 XRD patterns of MnO_x/TiO₂(n) catalysts: n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

3.4 Textural properties of the catalysts

Fig. 4 shows the N₂ adsorption/desorption isotherms and pore size distributions of the MnO_x/TiO₂(n) catalysts with various TiO₂ particle sizes. For all the MnO_x/TiO₂(n) catalysts, the N₂ adsorption/desorption isotherms show similar type IV behavior and hysteresis loops attributed to type H1. The corresponding quantification parameters of the textural properties, including BET specific surface area, total pore volume and average pore diameter, are shown in Table 1. All of the catalysts showed similar specific surface areas (50.2–57.4 m² g⁻¹), pore volumes (0.27–0.29 mL g⁻¹) and mean pore diameters (20–27 nm). It should be noted that the pore sizes were much larger than the dynamic size of the feed gas, indicating that the pore structures of the MnO_x/TiO₂(n) catalysts were feasible for reactant molecule diffusion.

Fig. 4 N₂ adsorption/desorption isotherms (A) and BJH pore distribution curves (B) of MnO_x/TiO₂(n) catalysts: n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

Table 1 BET specific surface area, total pore volume, mean pore diameter surface composition of MnO_x/TiO₂(n) catalysts with various TiO₂ support particle sizes

Catalyst	BET surface area (m^2 g^−1)	Pore volume (mL g^−1)	Average pore diameter (nm)	Mn^4+/Mntotal^a
a Determined by XPS.
MnOx/TiO2(5–10)	51.5	0.29	22.2	0.29
MnOx/TiO2(10–25)	57.4	0.29	27.2	0.48
MnOx/TiO2(40)	54.0	0.27	19.8	0.35
MnOx/TiO2(60)	50.2	0.29	23.4	0.34

---

3.5 H₂-TPR

The redox properties of catalysts play an important role in the NH₃-SCR reaction. Thus, H₂-TPR experiments were carried out to determine the oxidative properties of the MnO_x/TiO₂(n) catalysts and the results are shown in Fig. 5. All of the catalysts exhibited two broad hydrogen consumption peaks. The lower temperature reduction peak in the temperature range of 300–450 °C could be attributed to the reduction of Mn⁴⁺ to Mn³⁺ and Ti⁴⁺ to Ti³⁺, which originated from interaction between MnO_x and TiO₂.^23,38–41 The higher temperature peak in the temperature range of 450–600 °C could be ascribed to the reduction of Mn³⁺ to Mn²⁺.^38,42,43 It should be noted that the temperature peaks were different for MnO_x/TiO₂(n) with different initial TiO₂ particle sizes. For MnO_x/TiO₂(5–10), MnO_x/TiO₂(10–25), MnO_x/TiO₂(40) and MnO_x/TiO₂(60), the reduction peaks at lower temperature were centered at 400 °C, 364 °C, 384 °C and 367 °C, while the higher temperature reduction peaks were 525 °C, 516 °C, 527 °C and 521 °C, respectively. In previous studies, the SCR performance of MnO_x/TiO₂ catalysts has been related to the oxidation state of surface Mn atoms, especially the properties of surface Mn⁴⁺ species.^38,42,43 The lower peak temperature of MnO_x/TiO₂(n) represents a higher oxidative property and oxygen mobility. The order of the peak temperatures in the temperature range of 300–450 °C for the MnO_x/TiO₂(n) catalysts was as follows: MnO_x/TiO₂(10–25) (364 °C) < MnO_x/TiO₂(60) (367 °C) < MnO_x/TiO₂(40) (384 °C) < MnO_x/TiO₂(5–10) (400 °C). MnO_x/TiO₂(60) and MnO_x/TiO₂(40) showed different trends in their corresponding SCR catalytic activity. The catalytic activity was not only decided by the properties of active sites, but also relied on the number of active sites. The peak area in the temperature range of 300–450 °C for MnO_x/TiO₂(40), which was associated with the amount of Mn⁴⁺, was higher than that for MnO_x/TiO₂(60). Therefore, the properties and amount of Mn⁴⁺ in MnO_x/TiO₂(n) could also be affected by the TiO₂ particle size, thus influencing the SCR catalytic activity.

Fig. 5 H₂-TPR profiles of the MnO_x/TiO₂(n) catalysts: n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

3.6 O₂-TPD

In order to further confirm the oxidative properties and oxygen mobility, O₂-TPD was then used to characterize the oxygen mobility of the MnO_x/TiO₂(n) catalysts. Fig. 6 shows the O₂-TPD curves of the MnO_x/TiO₂(n) catalysts with various TiO₂ particle sizes. For all of the MnO_x/TiO₂(n) catalysts, there are two desorption peaks centered at 500–700 °C and 700–800 °C, respectively. The desorption of O₂ at a temperature above 200 °C in O₂-TPD for the manganese-based catalysts is associated with the evolution of oxygen from the lattice (surface or bulk).⁴⁴ The lower temperature peak in the temperature range of 500–700 °C was attributed to the surface lattice oxygen and related to the surface oxygen vacancies, which is often correlated with a high oxidation ability.^45,46 The higher temperature peak at temperatures above 700 °C could be assigned to bulk lattice oxygen, which was generally not correlated to oxidation ability and strongly bound to the lattice.⁴⁶ The peak temperature in the temperature range of 500–700 °C for MnO_x/TiO₂(10–25) was 560 °C, which was lower than that of MnO_x/TiO₂(5–10) (630 °C), MnO_x/TiO₂(40) (621 °C) and MnO_x/TiO₂(60) (586 °C). Meanwhile, the order of the lower temperature peak amount was similar to H₂-TPR.

Fig. 6 O₂-TPD profiles of MnO_x/TiO₂(n) catalysts of n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

3.7 NH₃-TPD

The surface acid sites of the catalysts played a significant role in the NH₃-SCR reaction, which could adsorb and activate NH₃. NH₃-TPD was used to study the surface acidity of MnO_x/TiO₂(n), as depicted in Fig. 7. For all the MnO_x/TiO₂(n) catalysts, the NH₃-TPD curves featured a broad peak between 100 and 400 °C. All these desorption peaks were divided into two or three peaks based on the Gaussian peak shape model. The desorption peaks were labeled as I, II and III, according to the peak temperature from low to high, which could be attributed to the NH₃ desorbed from weak acid sites, medium acid sites and strong acid sites, respectively.^17,47 The peak temperature and corresponding area in NH₃-TPD were calculated and are listed in Table 2. The peak temperature of MnO_x/TiO₂ with TiO₂ < 40 nm was ca. 200 °C, which is lower than that of MnO_x/TiO₂(60). More importantly, the order of the acid amount for the MnO_x/TiO₂(n) catalysts was as follows: MnO_x/TiO₂(10–25) > MnO_x/TiO₂(40) > MnO_x/TiO₂(60) > MnO_x/TiO₂(5–10), which is also in accordance with the catalytic activity of the counterpart catalysts, suggesting that the particle size of the support not only influences the oxidizing ability but also the acidic properties.

Fig. 7 NH₃-TPD profiles of the MnO_x/TiO₂(n) catalysts: n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

Table 2 Quantitative analysis of NH₃-TPD for the MnO_x/TiO₂(n) catalysts

Catalyst	Temperature (°C)			Acid amount (μmol g^−1)
	TI	TII	TIII	SI	SII	SIII	SI + SII + SIII
Mn/TiO2(5–10)	197	279	—	194	397	—	591
Mn/TiO2(10–25)	200	288	361	279	430	89	798
Mn/TiO2(40)	200	279	357	301	329	158	789
Mn/TiO2(60)	222	310	—	408	228	—	636

---

Compared with larger TiO₂ support particle size, a smaller TiO₂ particle size could provide a greater boundary area between active sites and supports, which is beneficial for the synergistic effect between acid sites and redox sites. As previously reported, low-temperature selective catalytic reduction mainly occurs via a Langmuir–Hinshelwood mechanism. The SCR activity was controlled not only by the activation of NO, O₂ and NH₃ reactant, but also by the synergistic effect between acid sites and redox sites.

3.8 XPS

In order to further acquire information regarding the oxidation state of manganese on the catalyst surface, XPS spectra of the MnO_x/TiO₂(n) catalysts were recorded. Fig. 8 shows the Mn 2p spectra of the MnO_x/TiO₂ catalysts with various TiO₂ particle sizes. The Mn 2p spectra of MnO_x/TiO₂(n) catalysts were deconvoluted based on the Gaussian–Lorentzian (GL(30)) line shape after subtracting a Shirley baseline. All the Mn 2p_1/2 and 2p_3/2 peaks of each Mn species have the same half height peak width and the center of the Mn 2p_1/2 and 2p_3/2 peaks was 11.7 eV. Mn 2p spectra of the MnO_x/TiO₂(n) catalysts could be divided into two peaks with maxima at 644.1–645.3 eV and 641.9–642.5 eV, which can be assigned to Mn⁴⁺ and Mn³⁺, respectively.^{23,41,42,48,49} The quantitation results of the Mn⁴⁺/Mn molar ratio in XPS spectra are shown in Table 1. It can be seen that the Mn⁴⁺/Mn molar ratio of the MnO_x/TiO₂(10–25) catalyst was 0.48, which is higher than 0.35 for MnO_x/TiO₂(40), 0.34 for MnO_x/TiO₂(60) and 0.29 for MnO_x/TiO₂(5–10). Moreover, the order of the surface Mn⁴⁺/Mn molar ratio on the MnO_x/TiO₂(n) catalysts was as follows: MnO_x/TiO₂(10–25) > MnO_x/TiO₂(40) > MnO_x/TiO₂(60) > MnO_x/TiO₂(5–10), which agrees well with the SCR catalytic performance. These results demonstrate that the TiO₂ particle size could influence the oxidation state of surface manganese oxides, leading to the variation in Mn⁴⁺ concentration, which could be ascribed to the interaction between manganese oxides and titanium.^42,50 Thirupathi et al. also investigated the effect of the surface manganese species on the SCR performance and found that surface Mn⁴⁺ on the TiO₂ support was directly associated with the SCR performance.⁴² All of these results revealed that initial TiO₂ particle size can affect the oxidizing ability and oxygen mobility. The higher oxidizing ability could increase the NO₂/NO_x ratio in the actual reaction feed, which could increase the proportion of “fast SCR” reaction route, and thereby enhance the SCR activity.⁵⁰

Fig. 8 XPS spectra for Mn 2p of the MnO_x/TiO₂(n) catalysts: n = 5–10 (a), 10–25 (b), 40 (c) and 60 (d).

3.9 In situ DRIFTS studies

3.9.1 NO + O₂ adsorption. In order to understand the mechanism of the MnO_x/TiO₂(n) catalysts, DRIFT spectra of MnO_x/TiO₂(10–25) were studied. Fig. 9 shows the DRIFT spectra of NO + O₂ adsorption over MnO_x/TiO₂(10–25) for various times at 100 °C. Bands at 1607, 1580, 1492, 1446, 1300, 1282 and 1257 cm⁻¹ were detected. The band at 1607 cm⁻¹ could be ascribed to the bridge nitrate.^50–53 The bands at 1580, 1300 and 1282 cm⁻¹ could be assigned to bidentate nitrate.^41,51–54 The bands at 1492 and 1257 cm⁻¹ could be identified as monodentate nitrate.^52,55 The wide peak in the region of 1430–1550 cm⁻¹ could be attributed to the bidentate nitrate on the surface of the sample.^52,55 The band at 1446 cm⁻¹ was assigned to nitro compounds.⁵⁴ In the first 5 min, only bridge nitrate (band at 1607 cm⁻¹) and bidentate nitrate (bands at 1580, 1300 and 1282 cm⁻¹) were detected on the manganese oxides. With increasing time, monodentate nitrate (bands at 1492 and 1257 cm⁻¹) on the manganese oxides appeared. Meanwhile, bidentate nitrate on the surface of the sample (the wide peak in the region of 1430–1550 cm⁻¹) increased with exposure time. These results illustrated that NO_x first adsorbed on manganese oxides as bridge and bidentate nitrate, and then as monodentate nitrate.

Fig. 9 DRIFT spectra of MnO_x/TiO₂(10–25) exposed to 500 ppm NO + 5% O₂ + N₂ at 100 °C for various times.

3.9.2 NH₃ adsorption after NO + O₂ adsorption. To understand the role of various adsorbed NO_x species during the NH₃-SCR reaction, the in situ DRIFT spectra for the pre-adsorbed NO + O₂ and NH₃ at 100 °C as a function of time are shown in Fig. 10. A reaction temperature of 100 °C was chosen to decrease the reaction rate; therefore the surface species on the catalyst surface could be enhanced to a detectable level.⁵⁶ When 500 ppm NH₃ was introduced to the catalyst pretreated by NO + O₂, the intensity of the band at 1607 cm⁻¹ assigned to bridge nitrate on manganese oxides decreased immediately with the increase of time. Meanwhile, the intensity of the bands at 1580 and 1295 cm⁻¹ (shown in Fig. 10C) assigned to bidentate nitrate on the manganese oxides decreased gradually. However, the variation of the intensity of the band at 1257 cm⁻¹ assigned to monodentate nitrate on the manganese oxides was different, and remained constant, indicating that the monodentate nitrate was not contributing to the SCR reaction at 100 °C. Simultaneously, the intensity of the bands at 3341 and 3244 cm⁻¹ (shown in Fig. 10D) assigned to asymmetric stretching and symmetric stretching of NH₃ coordinated to Lewis acid sites, increased immediately with the addition of NH₃.^51,57,58 It is reasonable to assume that the band at 1598 cm⁻¹ assigned to coordinated NH₃ on Lewis acid sites was caused by the overlap of the bridge and bidentate nitrates in the regions of 1630 and 1550 cm⁻¹. These results indicated that bridge and bidentate nitrates were the main contributors to the low-temperature SCR reaction, fitting the Langmuir–Hinshelwood mechanism suggested in previous works.^55,57,59,60 It should be noted that the band due to the bridge and bidentate nitrates did not completely vanish, even when the NH₃ purging time was prolonged to 60 min, suggesting that NO_x species adsorbed on various manganese species showed different SCR activity. More importantly, the band (shown in Fig. 10B) assigned to bridge nitrate shifted from 1607 cm⁻¹ to 1600 cm⁻¹, illustrating that adsorbed NO_x species at higher wavenumbers in the IR band showed superior SCR activity, since the IR band position was statistically evaluated according to the similar structures of the various surface manganese species. Based on our previous characterization results, such as those from H₂-TPR, O₂-TPD and XPS, the surface valence states of manganese with various initial support sizes contained different compositions of Mn⁴⁺ and Mn³⁺. The IR vibration frequencies in the N–O stretching modes could be varied via the formation of π-back bands.⁶¹ With high-valent cations (Mn⁴⁺), which are poor in d-electrons, the extent of π-back donation is lower than that in low-valent cations (Mn³⁺), thus increasing the vibration frequency or wavenumber. The bridge nitrate at a higher wavenumber could be attributed to the NO_x adsorbed on the high-valent manganese species, Mn⁴⁺. Based on the previous DRIFTS results, the higher wavenumber IR band (1607 cm⁻¹) decreased more quickly than the lower one. Therefore, we deduced that NO_x adsorbed species on Mn⁴⁺ was the dominant factor influencing the low-temperature SCR activity.

Fig. 10 DRIFT spectra of MnO_x/TiO₂(10–25) pretreated by exposure to 500 ppm NO + 5% O₂ + N₂ followed by exposure to 500 ppm NH₃ + N₂ at 100 °C for various times.

3.9.3 NO + O₂ adsorption after NH₃ adsorption. Furthermore, in order to understand the function of absorbed NH₃ species during the NH₃-SCR reaction, DRIFT spectra of MnO_x/TiO₂(10–25) in a flow of NO + O₂ were collected, after the catalyst was pre-exposed to a NH₃ for 60 min followed by N₂ purging for 30 min at 100 °C, as shown in Fig. S2.† The intensity of the bands at 3410 and 3244 cm⁻¹ assigned to NH stretching remained unchanged, illustrating that the low-temperature SCR reaction did not occur via an Eley–Rideal mechanism but via a Langmuir–Hinshelwood mechanism, which coincides with previous work.⁶

3.10 Origin of SCR activity

Combining the results obtained in this study, we could conclude that the low-temperature SCR activity over MnO_x/TiO₂ occurred via a Langmuir–Hinshelwood mechanism.

Therefore, the NH₃-SCR reaction over MnO_x/TiO₂ can be described by the following pathways.

At low temperature, the reaction pathway occurred as a Langmuir–Hinshelwood mechanism.⁶

O₂ (g) + * → 2O*

The type of adsorbed NO_x species on manganese oxides has a prominent effect on the SCR activity. The order of SCR activity of various adsorbed nitrates was as follows: bridge nitrate > bidentate nitrate > monodentate nitrate. The bridge nitrate and bidentate nitrate were the main contributors to the SCR activity, while the monodentate nitrate showed no activity, which agreed with previous work.⁵⁰ Moreover, among all the bridge nitrates on various manganese oxides, bridge nitrate on the high-valence manganese species (e.g. Mn⁴⁺) was the most active species in the low-temperature SCR reaction, which could been proved by the blue shift on the DRIFT spectra in a flow of NH₃ after pretreatment with NO + O₂. A higher Mn⁴⁺/Mn molar ratio could produce more bridge nitrate on Mn⁴⁺, thus increasing the SCR activity.

4 Conclusions

In this work, a series of MnO_x/TiO₂ catalysts with various initial TiO₂ particle sizes was prepared via an impregnation method for the selective catalytic reduction of NO with NH₃. The MnO_x/TiO₂(10–25) catalyst showed superior low-temperature SCR catalytic activity. The XRD results illustrated a high dispersion of manganese oxides on all the catalysts via interactions between manganese oxides and anatase TiO₂. H₂-TPR, O₂-TPD, NH₃-TPD and XPS results indicated that the acid amount and Mn⁴⁺/Mn molar ratio were dependent on the initial TiO₂ support particle size. The order of surface acidity and Mn⁴⁺/Mn molar ratio was as follows: MnO_x/TiO₂(10–25) > MnO_x/TiO₂(40) > MnO_x/TiO₂(60) > MnO_x/TiO₂(5–10), which agreed well with the order of SCR activity on the corresponding catalysts. Furthermore, the in situ DRIFTS results showed that the low-temperature SCR reaction occurred via a Langmuir–Hinshelwood mechanism and the adsorbed NO_x was the key factor in the low-temperature SCR activity. The order of adsorbed NO_x species activity on manganese oxides was as follows: bridge nitrate > bidentate nitrate > monodentate nitrate. Moreover, the bridge nitrate on Mn⁴⁺ was the greatest contributor to the low-temperature SCR activity. The variation of the initial TiO₂ support particle size influenced the surface Mn⁴⁺/Mn molar ratio of the MnO_x/TiO₂ catalysts, which would promote the reactivity of the bridge nitrate, therefore enhancing SCR performance.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support from the China Postdoctoral Science Foundation (2017M613143), the Natural Science Foundation of Shaanxi Province of China (2017JQ2016), and the National Natural Science Foundation of China (21802107).

References

1. X. Liu, Y. Zhang, W. Han, A. Tang, J. Shen, Z. Cui, P. Vitousek, J. W. Erisman, K. Goulding, P. Christie, A. Fangmeier and F. Zhang, Nature, 2013, 494, 459–462.
2. Y. Cheng, G. Zheng, C. Wei, Q. Mu, B. Zheng, Z. Wang, M. Gao, Q. Zhang, K. He, G. Carmichael, U. Poschl and H. Su, Sci. Adv., 2016, 2, e160153012.
3. S. Roy and A. Baiker, Chem. Rev., 2009, 109, 4054–4091.
4. Z. Hu, K. Q. Sun, W. Z. Li and B.-Q. Xu, Catal. Today, 2010, 158, 432–438.
5. Z. Hu, W.-Z. Li, K.-Q. Sun and B.-Q. Xu, Catal. Sci. Technol., 2013, 3, 2062–2071.
6. C. Liu, J. Shi, C. Gao and C. Niu, Appl. Catal., A, 2016, 522, 54–69.
7. R. Zhang, N. Liu, Z. Lei and B. Chen, Chem. Rev., 2016, 116, 3658–3721.
8. J. Wang, H. Zhao, G. Haller and Y. Li, Appl. Catal., B, 2017, 202, 346–354.
9. W. Shan and H. Song, Catal. Sci. Technol., 2015, 5, 4280–4288.
10. Z. Liu, J. Li and S. I. Woo, Energy Environ. Sci., 2012, 5, 8799–8814.
11. F. Liu, W. Shan, D. Pan, T. Li and H. He, Chin. J. Catal., 2014, 35, 1438–1445.
12. A. M. Beale, F. Gao, I. Lezcano-Gonzalez, C. H. F. Peden and J. Szanyi, Chem. Soc. Rev., 2015, 44, 7371–7405.
13. X. Hu, L. Huang, J. Zhang, H. Li, K. Zha, L. Shi and D. Zhang, J. Mater. Chem. A, 2018, 6, 2952–2963.
14. K. Zha, L. Kang, C. Feng, L. Han, H. Li, T. Yan, P. Maitarad, L. Shi and D. Zhang, Environ. Sci.: Nano, 2018, 5, 1408–1419.
15. K. Zha, S. Cai, H. Hu, H. Li, T. Yan, L. Shi and D. Zhang, J. Phys. Chem. C, 2017, 121, 25243–25254.
16. C. Li, X. Tang, H. Yi, L. Wang, X. Cui, C. Chu, J. Li, R. Zhang and Q. Yu, Appl. Surf. Sci., 2018, 428, 924–932.
17. Z. Fan, J. Shi, C. Gao, G. Gao, B. Wang and C. Niu, ACS Appl. Mater. Interfaces, 2017, 9, 16117–16127.
18. L. Yan, Y. Liu, K. Zha, H. Li, L. Shi and D. Zhang, ACS Appl. Mater. Interfaces, 2017, 9, 2581–2593.
19. P. G. Smirniotis, D. A. Pena and B. S. Uphade, Angew. Chem., Int. Ed., 2001, 40, 2479–2482.
20. P. G. Smirniotis, P. M. Sreekanth, D. A. Peña and R. G. Jenkins, Ind. Eng. Chem. Res., 2006, 45, 6436–6443.
21. Y. J. Kim, H. J. Kwon, I. Nam, J. W. Choung, J. K. Kil, H. Kim, M. Cha and G. K. Yeo, Catal. Today, 2010, 151, 244–250.
22. J. Li, J. Chen, R. Ke, C. Luo and J. Hao, Catal. Commun., 2007, 8, 1896–1900.
23. P. R. Ettireddy, N. Ettireddy, S. Mamedov, P. Boolchand and P. G. Smirniotis, Appl. Catal., B, 2007, 76, 123–134.
24. S. Yang, F. Qi, S. Xiong, H. Dang, Y. Liao, P. K. Wong and J. Li, Appl. Catal., B, 2016, 181, 570–580.
25. B. Xu, J. Wei, Y. Yu, J. Li and Q. Zhu, Top. Catal., 2003, 22, 77–85.
26. B. Xu, J. Wei, Y. Yu, Y. Li, J. Li and Q. Zhu, J. Phys. Chem. B, 2003, 107, 5203–5207.
27. Q. Zhang, Y. Li and B. Xu, Catal. Today, 2004, 98, 601–605.
28. M. M. Pakulska, C. M. Grgicak and J. B. Giorgi, Appl. Catal., A, 2007, 332, 124–129.
29. M. Murdoch, G. I. N. Waterhouse, M. A. Nadeem, J. B. Metson, M. A. Keane, R. F. Howe, J. Llorca and H. Idriss, Nat. Chem., 2011, 3, 489–492.
30. I. I. Soykal, H. Sohn and U. S. Ozkan, ACS Catal., 2012, 2, 2335–2348.
31. Z. Hu, S. Tan, R. Mi, X. Li, D. Li and B. Yang, Catal. Lett., 2018, 148, 1490–1498.
32. J. Li, H. Chang, L. Ma, J. Hao and R. T. Yang, Catal. Today, 2011, 175, 147–156.
33. M. Koebel, G. Madia and M. Elsener, Catal. Today, 2002, 73, 239–247.
34. M. Devadas, O. Krocher, M. Elsener, A. Wokaun, N. Soger, M. Pfeifer, Y. Demel and L. Mussmann, Appl. Catal., B, 2006, 67, 187–196.
35. E. F. Massoth, Adv. Catal., 1978, 2, 265–310.
36. G. C. Bond and S. F. Tahir, Appl. Catal., 1991, 71, 1–31.
37. J. M. Gallardo-Amores, T. Armaroli, G. Ramis, E. Finocchio and G. Busca, Appl. Catal., B, 1999, 22, 249–259.
38. E. Park, S. Chin, J. Jeong and J. Jurng, Microporous Mesoporous Mater., 2012, 163, 96–101.
39. R. Guo, Q. Wang, W. Pan, W. Zhen, Q. Chen, H. Ding, N. Yang and C. Lu, Appl. Surf. Sci., 2014, 317, 111–116.
40. S. S. R. Putluru, L. Schill, A. D. Jensen, B. Siret, F. Tabaries and R. Fehrmann, Appl. Catal., B, 2015, 165, 628–635.
41. W. Li, R. Guo, S. Wang, W. Pan, Q. Chen, M. Li, P. Sun and S. Liu, Fuel Process. Technol., 2016, 154, 235–242.
42. B. Thirupathi and P. G. Smirniotis, J. Catal., 2012, 288, 74–83.
43. S. M. Lee, K. H. Park, S. S. Kim, D. W. Kwon and S. C. Hong, J. Air Waste Manage. Assoc., 2012, 62, 1085–1092.
44. J. Luo, Q. Zhang, J. Garcia-Martinez and S. L. Suib, J. Am. Chem. Soc., 2008, 130, 3198–3207.
45. T. Seiyama, Catal. Rev., 1992, 34, 281–300.
46. M. Dhakad, S. S. Rayalu, R. Kumar, P. Doggali, S. Bakardjieva, J. Subrt, T. Mitsuhashi, H. Haneda and N. Labhsetwar, Catal. Lett., 2008, 121, 137–143.
47. S. Cimino, L. Lisi and M. Tortorelli, Chem. Eng. J., 2016, 283, 223–230.
48. Z. Wu, R. Jin, H. Wang and Y. Liu, Catal. Commun., 2009, 10, 935–939.
49. P. Sun, R. Guo, S. Liu, S. Wang, W. Pan and M. Li, Appl. Catal., A, 2017, 531, 129–138.
50. H. Hu, K. Zha, H. Li, L. Shi and D. Zhang, Appl. Surf. Sci., 2016, 387, 921–928.
51. B. Jiang, Z. Li and S. Lee, Chem. Eng. J., 2013, 225, 52–58.
52. S. Wang, R. Guo, W. Pan, Q. Chen, P. Sun, M. Li and S. Liu, Catal. Commun., 2017, 89, 143–147.
53. L. Ma, Y. Cheng, G. Cavataio, R. W. McCabe, L. Fu and J. Li, Appl. Catal., B, 2014, 156–157, 428–437.
54. Z. Wu, B. Jiang, Y. Liu, H. Wang and R. Jin, Environ. Sci. Technol., 2007, 41, 5812–5817.
55. L. Wei, S. Cui, H. Guo and L. Zhang, Comput. Mater. Sci., 2018, 144, 216–222.
56. M. Devadas, O. Kröcher, M. Elsener, A. Wokaun, G. Mitrikas, N. Söger, M. Pfeifer, Y. Demel and L. Mussmann, Catal. Today, 2007, 119, 137–144.
57. F. Liu and H. He, J. Phys. Chem. C, 2010, 114, 16929–16936.
58. L. Wei, S. Cui, H. Guo, X. Ma and L. Zhang, J. Mol. Catal. A: Chem., 2016, 421, 102–108.
59. E. Tronconi, I. Nova, C. Ciardelli, D. Chatterjee and M. Weibel, J. Catal., 2007, 245, 1–10.
60. H. Hu, S. Cai, H. Li, L. Huang, L. Shi and D. Zhang, ACS Catal., 2015, 5, 6069–6077.
61. H. I. Konstantin, Catal. Rev., 2000, 42, 71–144.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ra10077b

---