A review of Mn-containing oxide catalysts for low temperature selective catalytic reduction of NO_x with NH₃: reaction mechanism and catalyst deactivation

Abstract

Atmospheric pollutants of nitrogen oxides (NO_x) can be reduced by selective catalytic reduction (SCR). SCR of NO_x with ammonia (NH₃) at low temperatures has attracted much interest for high nitric oxide (NO) conversion, and this method is dominated by catalysts. Manganese (Mn)-containing oxide catalysts exhibit high activity and selectivity for the unique redox property of manganese oxides (MnO_x). The reaction mechanisms and deactivation processes are summarized in this review. SCR of NO_x with NH₃ follows both the Langmuir–Hinshelwood and the Eley–Rideal mechanisms, which also contribute to the nitrous oxide formation. Fast SCR has a higher reaction rate than standard SCR. Mn-containing catalysts could also be deactivated by sulfur oxides and water vapor. The deactivation process of sulfur dioxide can be classified into two categories: deposition of (NH₄)₂SO₄ and sulfation of active sites. The deactivation caused by water vapor can be attributed to the competitive adsorption. The adsorption of water on catalysts' surface blocked the active sites, which are provided for the adsorption of NH₃ and NO. Alkali, alkaline earth and heavy metal ions existing in fine fly ash can also damage the catalysts' acid sites. A notable improvement on performance was obtained when Mn-containing catalysts were doped with a transition metal, for these enhanced its adsorption capacity and oxidation ability. Furthermore, this review gives a comprehensive discussion of the synergistic mechanism between bi-metal or multi-metal oxides. Major conclusions and several possible directions for further research are presented finally.

---

1. Introduction

Nitrogen oxides (NO_x) are a series of active gases, and include nitrogen dioxide (NO₂), nitrogen oxide (NO) and nitrous oxide (N₂O), and so on. Human activities cause a huge emission rate of NO_x, which is double that of the biotic and abiotic nitrogen fixation rates. Released NO_x can cause a series of environmental issues, such as photochemical smog, acid rain, and ozone depletion, and it can affect global tropospheric chemistry.^1–4 Great efforts have been devoted to abating the emission of NO_x.

The technologies used to control NO_x emission can be categorized as combustion controls and post-combustion controls.^5,6 Combustion controls, which aim to control the production of NO_x, include low NO_x burners,⁷ air graded burning and staged fuel combustion.⁸ Post-combustion controls aim to decrease the NO_x produced by reducing active N to fixed nitrogen gas (N₂). The technologies for reducing NO_x from flue gas can be divided into: direct decomposition,^9,10 selective catalytic reduction (SCR),^11,12 selective non-catalytic reduction (SNCR),^13,14 hybrid SNCR/SCR¹⁵ and NO_x storage-reduction catalysis.¹⁶ With the advantages of high efficiency and low cost, NO_x emitted from stationary sources (e.g., thermal plants or industrial boilers) has been predominantly controlled by SCR of NO with ammonia (NH₃-SCR) in the presence of excess oxygen (O₂) for decades.¹⁷

The catalyst to be used is a decisive factor in the process of decreasing NO_x (deNO_x). The common catalysts include noble metal catalysts,¹⁸ metal-exchanged zeolite catalysts,¹⁹ metal oxide catalysts,^20,21 heteropoly acid catalysts,²² and so on. Metal oxide catalysts are widely applied in NH₃-SCR. Nowadays, the most widely used catalysts are vanadium(V)-based catalysts and tungsten trioxide (WO₃) and/or molybdenum trioxide (MoO₃) doped vanadium(V) oxide (V₂O₅)/titanium dioxide (TiO₂) catalysts. These are usually installed at the upstream of flue gas because they require a higher working temperature of 300–400 °C.²³ However, some tough problems have not been solved, such as the effect of excessive dust pollution to the catalysts upstream of the flue, the deactivation by sulfur dioxide (SO₂) and alkali metal ions, the poor thermal stability at high temperatures and the toxicity of vanadium from the disabled catalysts.²⁴ One of the efficient ways to overcome these obstacles is transferring the SCR reactor from upstream to downstream of the flue gas, where there is relatively less dust and sulfur oxides in the flue gas but a lower temperature below 300 °C.²⁵

A series of metal oxide catalysts have been investigated to adapting low temperature, such as cerium (Ce), cobalt (Co), copper (Cu), iron (Fe), manganese (Mn), molybdenum (Mo), nickel (Ni) and V.^26–30 Of these, manganese oxides (MnO_x) catalysts show a notable NO conversion and N₂ selectivity for its multi oxidation state, high valence state and characteristic crystallinity. Peña et al.²⁶ advocated that MnO_x/TiO₂ had the highest activity among Co, chromium (Cr), Cu, Fe, Mn, Ni and V oxides supported on TiO₂ at low temperatures. Manganese dioxide (MnO₂) and manganese(III) oxide (Mn₂O₃) show the highest activity and N₂ selectivity, respectively, among several MnO_x.³¹ The activity and poison tolerance can be improved by doping with other transition metals. Ceria (CeO₂) provides sufficient oxygen in the reaction of redox NO_x, and improves the activity of MnO_x catalysts.^32–34 Mn–Fe spinel shows an excellent SCR performance at low temperature.³⁵ Other Mn containing catalysts, such as MnO_x–CoO_x/TiO₂,^28,36 MnO_x–CrO_x/TiO₂,^37,38 MnO_x–CuO_x,^39,40 lanthanum manganite (LaMnO₃),⁴¹ have been investigated by many researchers. Mn containing catalysts have been recognized as the potential alternative for industrial applications.

To date, advances in low temperature NH₃-SCR of NO_x have been reviewed.^5,6,42 A review by Li et al.⁴³ summarized the use of metal oxides and zeolite catalysts and focused on the catalysts' components, preparation process and catalytic performance, however, the reaction mechanisms were not clarified clearly. A recent review in 2016 by Liu et al.⁴⁴ summarized the use of MnO_x-based catalysts and concentrated on the technological processes and improvement methods, however, little effort was made to summarize the reaction mechanisms and catalyst deactivation processes.

In this review, the advances in the use of Mn containing oxide catalysts are summarized. The focal point of this review is to address the reaction mechanisms and deactivation processes of Mn containing oxide catalysts. The N₂ selectivity and side reactions are discussed together. This review gives a comprehensive discussion of the synergistic effects between bi-metal or multi-metal oxides. The deactivation process using sulfur oxides, water vapor, alkali metal and heavy metal ions and the regeneration methods are summarized. Finally, the major conclusions and several possible directions of research are presented.

2. Reaction mechanisms

To meet the newest and stringent emission standards, (NO_x concentration ≤ 50 mg m⁻³),⁴⁵ academic researchers and engineers are more interested in use of low temperature SCR, which is one of the efficient ways to install a processor downstream of the flue. A number of metal oxide catalysts have been investigated so far. Transition metal oxides play an important role in low temperature SCR catalysts, such as V₂O₅, MnO_x, CeO₂ and copper oxide (CuO). Of these, MnO_x shows an excellent performance because of its different crystallinity, special surface area and multi oxidation. It is vital to elucidate the reaction mechanisms for future research. In this section, the reaction mechanisms of NH₃-SCR over Mn-containing oxide catalysts are summarized.

2.1 Standard SCR

The NH₃-SCR of NO aims to reduce active N to fixed N₂, which is harmless to the atmosphere. In the presence of excess O₂, the main overall reaction is eqn (1).⁴⁶ A great number of studies have proposed that eqn (1) shows the reaction stoichiometry in typical SCR conditions.^47–50 In the absence of O₂, reaction in eqn (1) would convert into the reaction in eqn (2):⁵¹

4NH₃ + 4NO + O₂ → 4N₂ + 6H₂O(g), ΔG⁰₂₉₈ = −1651 kJ mol⁻¹ (1)

4NH₃ + 6NO → 5N₂ + 6H₂O(g), ΔG⁰₂₉₈ = −1821 kJ mol⁻¹ (2)

Because the content of NO is more than 90% among NO_x, eqn (1) is proposed as the standard SCR reaction and dominates the reaction stoichiometry. It is reported widely that the NH₃-SCR of the NO reaction when comparing the stoichiometric conditions follows both the Langmuir–Hinshelwood (L–H) mechanism and the Eley–Rideal (E–R) mechanism.^52,53 Through the L–H mechanism, both NH₃ and NO are adsorbed on the surface of catalysts. However, via the E–R mechanism, adsorbed NH₃ reacts with gaseous NO. It is suggested that the gaseous NH₃ could be adsorbed on both Lewis acid sites and Brønsted acid sites, however, the gaseous NO is mainly adsorbed by a physical adsorption process.⁵⁴ The adsorption of NH₃ has been recognized as the first step of the SCR reaction because it is easier for NH₃ to be adsorbed on acid sites rather than NO, O₂ and the reaction products.⁵⁵

The SCR process over MnO_x catalysts via the L–H mechanism can be approximately described as follows:^23,53,56

NH₃(g) → NH₃(ad) (3)

NO(g) → NO(ad) (4)

Mnⁿ⁺=O + NO(ad) → Mn⁽ⁿ⁻¹⁾⁺–O–NO (5)

NH₃(ad) + Mn⁽ⁿ⁻¹⁾⁺–O–NO → Mn⁽ⁿ⁻¹⁾⁺–O–NO–NH₃ → Mn⁽ⁿ⁻¹⁾⁺–OH + N₂ + H₂O (6)

Mn⁽ⁿ⁻¹⁾⁺–OH + 1/4O₂ → Mnⁿ⁺=O + 1/2H₂O (7)

Eqn (3) and (4) are the adsorption of gaseous NH₃ and NO. NH₃ is usually adsorbed on the Lewis acid sites and Brønsted acid sites to form adsorbed NH₃ species of coordinated NH₃ and ionic NH₄⁺, respectively.⁵⁷ Nevertheless, the coordinated NH₃ on the Lewis acid sites possesses a higher thermal stability than the ionic NH⁴⁺ on Brønsted acid sites. Manganese cations can provide a great number of Lewis acid sites.^49,58

Fang et al.^59,60 investigated the adsorption of NH₃ on the Mn₂O₃ (222), manganese(II,II) oxide (Mn₃O₄) (211) and MnO₂ (110) surfaces using density functional theory. It is claimed that, with more negative adsorption energy values and the shorter Mn–N bonds, Mn₂O₃ (222) and Mn₃O₄ (211) surfaces were more active for NH₃ adsorption than the MnO₂ (110) surface, which contributed to a higher performance (Table 1). Kapteijn et al.³¹ proposed that the highest NO conversion is exhibited by MnO₂, followed by Mn₅O₈, Mn₂O₃ and Mn₃O₄.

Table 1 The NO conversion of pure MnO_x^59a

MnOx	NO conversion (%)
Reaction temperature	353 K	373 K	393 K	413 K	433 K
a Reaction conditions: [NO] = 720 ppm, [NH3] = 800 ppm, [O2] = 3%. (Reprinted with permission from ref. 59. Copyright 2013 Elsevier.)
MnO2	13	14	16	19	21
Mn2O3	14	17	37	47	44
Mn3O4	18	23	34	44	56

---

The adsorbed NO is oxidized by the high valency state Mnⁿ⁺ cations, (e.g., Mn⁴⁺) on the catalysts' surface to form adsorbed monodentate nitrite (Mn⁽ⁿ⁻¹⁾⁺–O–NO) and the very metal cations are reduced as Mn⁽ⁿ⁻¹⁾⁺ [eqn (5)]. Furthermore, Mn⁽ⁿ⁻¹⁾⁺–O–NO reacts with adsorbed NH₃ species to form Mn⁽ⁿ⁻¹⁾⁺–O–NO–NH₃, which decomposes subsequently to N₂ and water (H₂O) [eqn (6)]. Then, the reduced Mn⁽ⁿ⁻¹⁾⁺ ions are regenerated by gaseous O₂ [eqn (7)].

The SCR process over MnO_x catalysts via the E–R mechanism can be described approximately as follows:^35,48,61

NH₃(g) → NH₃(ad) (8)

NH₃(ad) + Mnⁿ⁺=O → NH₂(ad) + Mn⁽ⁿ⁻¹⁾⁺–OH (9)

NH₂(ad) + NO(g) → NH₂NO → N₂ + H₂O (10)

Mn⁽ⁿ⁻¹⁾⁺–OH + 1/4O₂ → Mnⁿ⁺=O + 1/2H₂O (11)

The adsorption of NH₃ on the Lewis acid sites is recognized as the first step of NO reduction via the E–R mechanism. Coordinated NH₃ could be deprived of a hydrogen and be activated by the labile oxygen or the lattice oxygen of metal oxides to form an amine (NH₂) species [eqn (9)]. Labile oxygen can be released via the change of the valence states of Mn. Activated NH₂ species on the catalysts' surface reacted with gaseous NO to form the most important intermediate of NH₂NO, which subsequently decomposes to N₂ and H₂O [eqn (10)]. Then, the reduced Mn⁽ⁿ⁻¹⁾⁺ cations could be oxidized by O₂.

Furthermore, the formation of NH₄NO₂ is a typical SCR mechanism for Mn-containing catalysts. Qi and Tang,⁵⁶ and Eigenmann et al.⁶² proposed an amide–nitrosamine type mechanism, which is actually similar to the E–R mechanism. An extra species of NH₄NO₂ was presented in this mechanism. NH₄NO₂ could be decomposed to NH₂NO and H₂O, and is then decomposed to N₂ and H₂O [eqn (12)–(14)]:

OH(ad) + NO₂(ad) → O(ad) + HNO₂(ad) (12)

NH₃(ad) + HNO₂(ad) → NH₄NO₂(ad) → NH₂NO(ad) + H₂O (13)

NH₂NO(ad) → N₂ + H₂O (14)

In accordance with the transient eqn (3)–(11), Mn³⁺–O–NO–NH₃ and NH₂NO are the most important intermediate in the reaction of the L–H mechanism and E–R mechanism, respectively. There is a quite similarity between these two different mechanisms. A comproportionation, (i.e., N³⁺ and N³⁻, N²⁺and N²⁻) occurs on both the L–H and E–R mechanism (eqn (6) and (10)).^53,63

2.2 Fast SCR

A fast SCR reaction of NH₃ with NO + NO₂ over Mn-containing oxide catalysts has been reported. It is suggested that the fast SCR has a higher reaction rate than standard SCR.⁶⁴ Fast SCR was firstly investigated by Koebel et al., and Madia et al.^65–67 The general reaction can be described as follows:^68,69

4NH₃ + 2NO₂ + O₂ → 3N₂ + 6H₂O(g), ΔG⁰₂₉₈ = −1412 kJ mol⁻¹ (15)

4NH₃ + 2NO + 2NO₂ → 4N₂ + 6H₂O, ΔG⁰₂₉₈ = −1581 kJ mol⁻¹ (16)

In the presence of O₂, NO can be oxidized by active oxygen to form NO₂ [eqn (17)].⁷⁰ Judged by the Gibbs free energy, the reaction shown in eqn (15) does not occur easily and consequently limits the rate of eqn (15) or (16). Mn-containing metal oxide catalysts could catalyze this reaction in some extent:^71,72

2NO + O₂ → 2NO₂, ΔG⁰₂₉₈ = −70 kJ mol⁻¹ (17)

NO₂ is the difference between fast SCR standard SCR. NO₂ acts as a more efficient oxidizing agent than O₂ in the redox process of the SCR reaction. NO₂ can form surface nitrites and nitrates via dimerization:⁷³

2NO₂ → N₂O₄ (18)

N₂O₄ + H₂O → HNO₂ + HNO₃ (19)

NH₄NO₃ is formed by the reaction between NH₃ and HNO₃. NH₄NO₃ or its related surface species is the key intermediate in the fast SCR process. The reaction processes can be described as follows:

2NH₃ + 2NO₂ → N₂ + NH₄NO₃ + H₂O (20)

NH₄NO₃ + NO → N₂ + NO₂ + 2H₂O (21)

Many researchers considered that NH₄NO₃ would be solid below 170 °C. NH₄NO₃ could be reduced by NO at a higher temperatures [eqn (21)].^64,74 It is pointed out that NH₃ can restrain fast SCR by inhibiting the formation of NO₂ at 150–170 °C.⁷⁵ Actually, eqn (21) can be described as two intermediate reactions:

NH₄NO₃ ↔ NH₃ + HNO₃ (22)

2HNO₃ + NO → 3NO₂ + H₂O (23)

There is a chemical equilibrium in the fast SCR process [eqn (22)]. The formation of HNO₃ will be restrained while the NH₃ concentration is raised, and that inhibits the formation of NO₂ [eqn (23)]. Among the fast SCR processes, the vital process is the redox reaction between NO and HNO₃, which dominates the rate of fast SCR.

The performance of low temperature SCR has been extensively investigated. Excellent NO conversion and N₂ selectivity has been observed using simulated flue gas in the laboratory. Qi and Yang⁷⁶ obtained more than 99% of NO conversion on the MnO_x(0.3)–CeO₂ catalyst sintered at 120 °C. Long et al.⁷⁷ investigated the Fe–Mn-based catalysts. These catalysts showed nearly 100% NO conversion at 100–180 °C. Recently, France et al.⁷⁸ studied the CeO₂ modified FeMnO_x catalysts, and more than 95% NO conversion was obtained at 90–135 °C without the influence of SO₂ and H₂O. Zhu et al.⁷⁹ studied the holmium (Ho) modified Fe–Mn/TiO₂ catalysts, which revealed good performance for NO conversion and high SO₂ tolerance. However, more attempts need to be made to understand the fundamental mechanism of low temperature SCR, such as surface chemistry, crystal structure, kinetics and scientific reaction mechanism. These have a great influence on the performance of catalysts and knowledge of them would be beneficial in designing a new catalyst.

2.3 Side reactions

As the reductant, NH₃ is a vital resource in the SCR reaction. NH₃ is consumed mainly via N₂ and N₂O formation and the oxidation of the catalyst to NO_x.⁵² The wastage of NH₃ is a huge additional cost of the deNO_x process. To decrease the wastage of NH₃, an appropriate NH₃/NO ratio is necessary. Authors agree that a NH₃/NO ratio near to 1 is good. Furthermore, undesired reactions can occur during the SCR process. Eqn (24) and (25) show the undesired ammonia loss:

4NH₃ + 4O₂ → 2N₂O + 6H₂O(g), ΔG⁰₂₉₈ = −1102 kJ mol⁻¹ (24)

4NH₃ + 3O₂ → 2N₂ + 6H₂O(g), ΔG⁰₂₉₈ = −1310 kJ mol⁻¹ (25)

These are the thermodynamically favored reactions but they occur rarely in practice.⁸⁰ In addition, there is another undesired reaction during the NH₃-SCR process:

4NH₃ + 5O₂ → 4NO + 6H₂O, ΔG⁰₂₉₈ = −964 kJ mol⁻¹ (26)

Wang et al.⁸¹ claim that eqn (26) may replace eqn (1) as the dominant reaction over MnO_x/TiO₂ catalysts when the temperature was raised higher than 175 °C. This was proved by the determination of the components of outlet flue gas. This oxidization of NH₃ gives a decline in NO conversion and extra consumption of NH₃.

When the concentration of NH₃ is appropriate, the formation of N₂O is the primary waste of NH₃ and this decreases the N₂ selectivity [eqn (27)].³⁵

4NH₃ + 4NO + 3O₂ → 4N₂O + 6H₂O(g), ΔG⁰₂₉₈ = −1240 kJ mol⁻¹ (27)

Adsorbed NH₃ is oxidized on the catalyst surface to form an amine species (NH₂), which subsequently reacts with NO to form N₂ and H₂O. However, when a further hydrogen atom is abstracted from NH₂ to form an NH species, a N₂O species will be formed by the reaction of the NO and NH species.⁸² Both the L–H mechanism and the E–R mechanism pathways contribute to N₂O formation.

As previously mentioned, in the L–H mechanism, physically adsorbed NO can be oxidized by Mnⁿ⁺ to Mn⁽ⁿ⁻¹⁾⁺–O–NO, which can be further oxidized to monodentate nitrate (Mn⁽ⁿ⁻¹⁾⁺–O–NO₂) [eqn (28)]. The Mn⁽ⁿ⁻¹⁾⁺–O–NO₂ can react with adsorbed NH₃ to form Mn⁽ⁿ⁻¹⁾⁺–O–NO₂–NH₃. Subsequently, Mn⁽ⁿ⁻¹⁾⁺–O–NO₂–NH₃ will be decomposed to N₂O [eqn (29)]:^17,83

Mn⁽ⁿ⁻¹⁾⁺–O–NO + (1/2)O₂ → Mn⁽ⁿ⁻¹⁾⁺–O–NO₂ (28)

Mn⁽ⁿ⁻¹⁾⁺–O–NO₂ + NH₃(ad) → Mn⁽ⁿ⁻¹⁾⁺–O–NO₂–NH₃ → Mn⁽ⁿ⁻¹⁾⁺–OH + N₂O + H₂O (29)

As previously mentioned in Section 2.2, the reaction of NH₄NO₃ with NO is a vital step in the fast SCR process. Zhu et al.⁷⁴ speculated that NH₄NO₃ could be decomposed to N₂O and H₂O via the L–H mechanism [eqn (30)]. Referring to eqn (28) and (29), the formation of N₂O could be attributed to the better capacity for NH₃ activation and adsorbed active nitrate species.

NH₄NO₃ → N₂O + 2H₂O (30)

As mentioned previously, in the E–R mechanism, NH₂ species can react with gaseous NO to form N₂ and H₂O. While the NH₂ species is further oxidized on the metal cation to NH species, N₂O will be formed by the reaction of the NH species and gaseous NO [eqn (31) and (32)].^55,63,84 It is obvious that the formation of NH₂NO is a crucial step of NO reduction, which is directly related to the NO conversion and N₂ selectivity.⁵⁸

NH₂ + Mnⁿ⁺=O → NH + Mn⁽ⁿ⁻¹⁾⁺–OH (31)

NH + NO(g) → N₂O + H⁺ (32)

Whether the adsorbed NO is oxidized to monodentate nitrate or the NH₂ species is dehydrated to NH, the N₂ selectivity will be restrained and N₂O is formed.⁸⁵ This is an important difference from the standard SCR. The formation of N₂ and N₂O during the SCR process is illustrated in Fig. 1.

Fig. 1 The scheme of the SCR reaction through L–H and E–R mechanisms over Mn–Fe spinel catalyst. (Reprinted with permission from ref. 35. Copyright 2014 American Chemical Society.)

Hinted at by the previous equations, it is obvious that the two N atoms in N₂O originate from NO and NH₃, respectively. Suárez et al.⁸⁶ pointed out that N₂O did not primarily originate from the NH₃ oxidation reaction. The feasible main reaction path is that between the coordinated NO₃⁻ (generated from NO/NO₂ in the presence of O₂) and the adsorbed NH_x. Tang et al.⁶³ demonstrated that the N₂O selectivity of the SCR reaction over β-MnO₂ was higher than that over α-Mn₂O₃ at 150 °C. The N₂O is generated directly from the reaction of NO with NH₃ via the E–R mechanism. Use of calcium (Ca) modification improves the performance of N₂ selectivity for Mn-containing catalysts.⁸⁷

It is suggested that N₂O formation mainly resulted via the E–R mechanism.⁵³ Yang et al.³⁵ studied the mechanism of N₂O formation over Mn–Fe spinel catalysts. N₂O formation via the E–R mechanism was much more than via the L–H mechanism over the Mn–Fe spinel catalysts. In addition, N₂O selectivity was not promoted by increasing the NO concentration, but it was increased with the increase in NH₃ concentration. N₂O selectivity is also related to the gas hourly space velocity (GHSV). It was also found that N₂O in the SCR reaction over Mn–Ce catalysts was generated via the E–R mechanism, not the L–H mechanism.⁸⁸ The choice of E–R or L–H mechanisms ways will vary with the changes of temperature. It is reported that the L–H mechanism plays the main role below 150 °C, and the E–R mechanism way dominates the SCR reaction at higher temperatures.^55,89

2.4 Synergistic effect

A pure metal oxide may not be suitable for practical applications because of its defects. However, the property of one metal oxide can be improved by introducing foreign metal cations into its lattice. There will also be an interaction between different metal oxides. For example, reports in the literature indicate that Mn–Ce mixed oxide catalysts demonstrated the best performance among a multitude of metal oxide catalysts. Ceria can enhance the adsorption of NO and O₂, which benefits the oxidation of NO to NO₂ and improve sulfur resistance. Qi and Tang⁷⁶ found that the oxidation of NO to NO₂ was increased significantly after addition of ceria to MnO_x and that it speeded up the overall process. Actually, pure CeO₂ cannot be applied in industry because of its small specific surface area and low thermal stability.⁹⁰ Meanwhile, as is reported,⁹¹ modification with titanium (Ti) or tin (Sn) can improve the SCR property of cerium oxides. Qi et al. and Imamura et al.^27,92 found using X-ray diffraction patterns that there was no manganese oxide phase in the calcined Mn–Ce catalyst prepared by a co-precipitation method. This indicated that strong interactions exist between manganese and cerium oxides, because Mn₂O₃ and MnO₂ can be detected in pure manganese oxide calcined at the same temperature.

The redox property of catalysts is the key factor of the NH₃-SCR processes.²⁹ Electronic transfer, showing as oxidation and reduction, plays quite an important role in catalytic reactions. The redox couples exist over the metal oxide catalysts, such as Mn⁴⁺/Mn³⁺, Ce⁴⁺/Ce³⁺ and Fe³⁺/Fe²⁺, which provide the redox cycles with excess oxygen. The activity of bi-metal and multi-metal oxide catalysts could be promoted by dual redox cycles. The general formula can be described as follows:

Mⁿ⁺ + N^m+ ↔ M⁽ⁿ⁻¹⁾⁺ + N^(m+1)+ (33)

There is a typical SCR reaction process via the E–R mechanism on Mn–Ce/TiO₂ and Mn–Ce/aluminium oxide (Al₂O₃) catalysts.^27,93 Manganese oxides and ceria oxides also interact. They can form a solid solution because of the similarity of their structure.⁹⁴ Ceria has a superior oxygen storage performance. Thus, the process of oxidizing Mn³⁺ to Mn⁴⁺ is enhanced by using ceria.⁹⁵

Liu et al.⁹⁶ investigated a Mn–Ce–Ti mixed oxide catalyst prepared using a hydrothermal method, and found that there were dual redox cycles, such as Mn⁴⁺ + Ce³⁺ ↔ Mn³⁺ + Ce⁴⁺ and Mn⁴⁺ + Ti³⁺ ↔ Mn³⁺ + Ti⁴⁺. These dual redox cycles can promote each other and facilitate the electron transfer between Mn, Ce and Ti active sites by decreasing the migration energy. The proposed schemes are as follows (Fig. 2).

Fig. 2 The scheme of dual redox cycle during SCR process. (Reprinted with permission from ref. 96. Copyright 2014 American Chemical Society.)

The scheme shows that Mn cation sites may be the main active site for the adsorption of N. Furthermore, the addition of Ce, Fe, Cu, Ni and so on, may show a synergistic effect, which facilitates the generation of Mn⁴⁺ from Mn³⁺. Kwon et al.⁹⁷ studied the MnO_x/CeO₂–TiO₂ catalyst system. When Ce was added to Mn/Ti, an oxygen bridge of Mn–O–Ce was formed and, thus enhanced the binding between Mn and O₂. This oxygen bridge provided a channel for the electron transfer between manganese and cerium cations, and particularly accelerated the oxidation of Mn³⁺ to Mn⁴⁺ by Ce⁴⁺.²³

Mn₂O₃ + 2CeO₂ → 2MnO₂ + Ce₂O₃ (34)

Among the Mn–Fe mixed oxide catalysts, electronic transfer occurs between the different oxidation states of Fe³⁺, Fe²⁺, Mn⁴⁺ and Mn³⁺.⁹⁸ The performance of the Mn/TiO₂ catalyst was improved by the addition of Fe.⁹⁹ The process can be described approximately as follows:

Fe³⁺ + Mn³⁺ ↔ Fe²⁺ + Mn⁴⁺ (35)

NO + Mn⁴⁺ → NO⁺(ad) + Mn³⁺ (36)

1/2O₂ + Fe²⁺ → Fe³⁺ + O⁻(ad) (37)

NO⁺(ad) + O⁻(ad) → NO₂ (38)

≡Fe²⁺ + ≡Mn⁴⁺ → ≡Fe³⁺ + ≡Mn³⁺ (39)

Liu et al.^100,101 investigated a series of WO₃-doped Mn–zirconium (Zr) mixed oxide catalysts. Using catalyst performance measurements, the SCR performance and poisoning tolerance of the Mn–Zr catalyst doped with WO₃ was higher than that for the Mn–Zr catalyst alone. There were redox couples of Mn⁴⁺/Mn³⁺ and W⁶⁺/W⁵⁺, (i.e., W⁵⁺ + Mn⁴⁺ ↔ W⁶⁺ + Mn³⁺). The redox property and the electron transfer was improved using these dual redox couples (Fig. 3). Thus, the electron transfer between Mn and W active sites was promoted and this contributes to the activation of NH₃ and an improvement of the NO conversion (Fig. 4).

Fig. 3 The electron transfer of redox couples of Mn⁴⁺/Mn³⁺ and W⁶⁺/W⁵⁺.

Fig. 4 NO_x conversion over MnZr and WMnZr catalysts at 300 °C. Reaction conditions: [NO] = [NH₃] = 500 ppm, [O₂] = 5%, [H₂O] = 5%, [SO₂] = 50 ppm, GHSV = 128 000 h⁻¹. (Reprinted from ref. 100. Copyright 2015, with permission from Elsevier.)

Metal oxides could catalyze the reduction of NO with NH₃ via the transfer of electrons.^102,103 As is known, catalysts play a role in accelerating the reaction rate. Referring to Fig. 2, it can be seen that metal cations provide the adsorption sites and function as the transfer station of electrons in the SCR process. Manganese mainly acts as the adsorption center for nitrogen. Mn⁴⁺ receives an electron from NO or NH₃ and will be reduced into Mn³⁺. Then the reduced Mn³⁺ would be restored to Mn⁴⁺ by an extra oxygen and then the next redox cycle starts. However, a faster pathway is via the transfer of an electron between metal oxides, such as Ce, Fe, W and so on. Therefore, to design a catalyst, it is necessary to introduce an element for the role of the adsorption and oxidation of nitrogen. Simultaneously, another element is required for superior oxygen storage to quickly restore the reduced element. The coordination of these two types of elements will improve the performance of SCR.

It is essential to characterize the catalysts' structure in order to design an excellent catalyst. The current technology for treating the exhaust gas is supported vanadium-based catalysts on TiO₂ modified by W or Mo addition. Depending on the coverage, different polymeric vanadium oxides (VO_x) could segregate at the surface and these exhibited different turnover frequency (TOF) and selectivity.^104,105 This could be interesting if the same trend existed for MnO_x species, however, there has been little research proposed on use of different polymeric MnO_x corresponding to their different performances. Ettireddy et al.¹⁰⁶ studied TiO₂ supported manganese oxide catalysts. Different TOFs were obtained on the Mn/TiO₂ loaded with different amounts of manganese (Table 2). It was proposed that the polymeric or microcrystalline form of MnO_x was envisaged at higher loadings. As a general trend, the TOF and selectivity decreased with the polymeric form increasing at higher loadings. However, further study should be done to confirm which kind of polymeric manganese was formed and its TOF and selectivity should also be determined.

Table 2 Brunauer–Emmett–Teller (BET) surface area and TOF of different loadings of Mn/TiO₂ catalysts^106a

Catalyst	BET surface area (m^2 g^−1)	TOF at different GHSV (h^−1)
		50 000	100 000
a Reprinted with permission from ref. 106. Copyright 2007 Elsevier.
5% Mn/TiO2	238	139.5	127.8
11.1% Mn/TiO2	229	60.1	58.7
16.7% Mn/TiO2	196	32.3	31.7
20% Mn/TiO2	183	27.8	27.5
24% Mn/TiO2	165	22.0	21.3

---

In this section, the reaction mechanisms have been summarized. It was supposed originally that NO₂ could be the reactant of SCR process. However, it is widely agreed that the main reactant for the SCR process is NO, while NO₂ is reduced by the fast SCR process.¹⁰⁷ The synergistic effect among the different metal cations is essential to improve the catalysts' performance, such as NO conversion, selectivity and poisons' tolerance. According to various reports, the low resistance to different poisons is the greatest obstacle for the application of low temperature SCR catalysts.

3. Catalyst deactivation

Because of the demands of high temperature operation, conventional SCR catalysts suffer a huge amount of damage from the sulfur oxides, water vapor, heavy metal ions and alkali and alkaline earth metal ions in the upstream of the flue gas.¹⁰⁸ Installing the reactor downstream of the desulfurizer and precipitator is an excellent way to avoid deactivation. Many metal oxide catalysts have been reported as being low temperature SCR catalysts,^99,109,110 however, commercial low temperature SCR catalysts have narrow fields of application because they are not immune to the residual SO₂ and H₂O contained in real flue gas. The poor tolerance of SO₂ and H₂O has been a major obstacle for practical applications.¹¹¹ Therefore, it is significant to illuminate the poisoning mechanisms of SO₂, H₂O and so on.

3.1 SO₂ and H₂O

Sulfur oxides are mainly generated from the combustion of fossil fuels and the sintering of ore. Residual SO₂ after desulfurization can still damage the metal oxide catalysts. The deposition of ammonium sulfates, such as ammonium bisulfate (NH₄HSO₄) and ammonium sulfate [(NH₄)₂SO₄], is the primary cause for the deactivation of metal oxide catalysts at low temperature.¹¹² The decomposition temperature of ammonium sulfite [(NH₄)₂SO₃] and (NH₄)₂SO₄ salts is higher than the operation temperature of the catalysts. Most researchers regard the poisoning of SO₂ as a major problem. The deactivation of SO₂ can be classified into two categories: deposition of (NH₄)₂SO₄ and sulfation of active sites. The undesired metal sulfates and (NH₄)₂SO₄ would occupy active sites on the surface and gradually deactivate the catalyst. The deactivation caused by water vapor can contribute to the competitive adsorption. The adsorption of H₂O on the catalysts' surface blocks the active sites, which are provided for the adsorption of NH₃ and NO.

3.1.1 Deposition of ammonium sulfates. The harm caused by (NH₄)₂SO₃ and (NH₄)₂SO₄ is to mainly block the active sites. The micropore surface area and volume was decreased after SO₂ was introduced in to a simulated flue gas.¹¹³ When excess O₂ exists in the flue gas, the trace residual SO₂ can be oxidized to SO₃, a reaction catalyzed by the metal active sites [eqn (34)]. Furthermore, it was proved that the SO₂ could be easily oxidized on the MnO_x catalysts' surface. Also, NO_x would further facilitate the oxidation of SO₂ to SO₃ [eqn (35)].¹¹¹ The reaction could be described approximately as follows:

SO₂ + 1/2O₂ → SO₃ (40)

NO₂ + SO₂ → NO + SO₃ (41)

Gaseous NH₃ was assisted by the Brønsted acid sites to form NH₄⁺, which could react with SO₂ or SO₃ to form (NH₄)₂SO₃ or (NH₄)₂SO₄, respectively. In addition, NH₄HSO₄ species were also generated in the flue gas. The formation of NH₄HSO₄, (NH₄)₂SO₃ and (NH₄)₂SO₄ can be described as follows:¹¹⁴

SO₃ + H₂O → H₂SO₄ (42)

H₂SO₄ + NH₃ → NH₄HSO₄ (43)

2NH₃ + SO₂ + H₂O → (NH₄)₂SO₃ (44)

2NH₃ + SO₃ + H₂O → (NH₄)₂SO₄ (45)

Actually, (NH₄)₂SO₃ and (NH₄)₂SO₄ can be decomposed at a relatively higher temperature. However, low temperature SCR of NO_x is usually requested at a low operation temperature, which is lower than the decomposition temperature of the (NH₄)₂SO₃ and (NH₄)₂SO₄. Therefore, removing the undesired side-products of (NH₄)₂SO₄ salts is a big challenge to researchers.

Almost all of reported MnO_x catalysts were affected by the introduction of SO₂ in the feed gas.¹¹⁵ Zhang et al.¹¹⁶ introduced 100 ppm SO₂ in the feed gas, which induced an apparent decrease of NO conversion over the Mn–Ce metal oxide catalysts supported on carbon nanotubes. Lu et al.¹¹⁷ fed 200 ppm SO₂ to the flue gas, and then the NO_x conversion of Mn–Ce/TiO₂ catalyst decreased from an initial value of 99% to about 78%. Jiang et al.¹¹⁸ investigated the effect of SO₂ on MnO_x(0.4)/TiO₂ catalysts prepared by three methods, sol–gel, impregnation and co-precipitation. The NO conversions had an apparent decrease for these catalysts (Fig. 5a).

Fig. 5 The effect of SO₂ on NO conversion. (a) Reaction conditions: [NO] = [NH₃] = 1000 ppm, [O₂] = 3%, [SO₂] = 200 ppm, balance N₂, temperature: 150 °C, GHSV = 30 000 h⁻¹. (Reprinted from ref. 118. Copyright 2008, with permission from Elsevier.) (b) Reaction conditions: [NO] = 600 ppm, [NH₃] = 480 ppm, [O₂] = 2%, [SO₂] = 300 ppm, [H₂O] = 10 vol%, balance N₂, temperature: 240 °C, GHSV = 24 000 h⁻¹. (Reprinted from ref. 111. Copyright 2009, with permission from Elsevier.)

Yu et al.¹¹¹ prepared MnO₂–Fe₂O₃–CeO₂–TiO₂ catalysts. The performance of this catalyst was decreased by introducing SO₂. The NH₄⁺ species and the SO₄²⁻ species were determined from Fourier-transform infrared spectra. The NH₄⁺ species were chemisorbed on to the Brønsted acid sites.¹¹⁹ This means that the poisoning of SO₂ can be via the formation and deposition of (NH₄)₂SO₄, which blocks the active channels of the catalyst. The NO conversion was decreased to 50% from 90% (Fig. 5b).

Xu et al.¹²⁰ also found that NH₄HSO₃ and NH₄HSO₄ formed via the reaction of SO₂ and NH₃ could be deposited on catalysts' surface and blocked the active sites [eqn (34)–(39)]. Furthermore, more Brønsted acid sites will be generated while the sulfates are formed by SO₂ adsorption on surface. The Lewis acid site could be transformed to the Brønsted acid site by adsorption of a water molecule.¹²¹ This means that a wet atmosphere would promote the formation of the Brønsted acid sites, which facilitates the sorption of NH₄⁺.¹²² In terms of diffuse reflectance infrared Fourier transform (DRIFT) spectra, Jiang et al.¹²³ proved that the formation of NH₄⁺ was promoted after introducing SO₂. However, even though Brønsted acid sites were formed by the sulfatization, NO conversion was decreased because SO₂ occupied the NO adsorption sites.

Therefore, to obtain high NO conversion, it is necessary to prevent the formation of (NH₄)₂SO₄. Actually, it is nearly impossible to eliminate the residual SO₂ completely. Efficient ways to do it may be preventing the oxidation of SO₂ and decreasing the decomposition temperature of (NH₄)₂SO₄ and NH₄HSO₄ on the catalysts' surface.

Jin et al.²⁵ studied the Mn–Ce/TiO₂ and Mn/TiO₂ catalysts. In terms of the thermogravimetry/differential scanning calorimetry (TG/DSC) results, the decomposition temperatures of (NH₄)₂SO₄ and NH₄HSO₄ on the Mn/TiO₂ catalyst was determined to be 213 °C and 361 °C, respectively. However, in the case of the Mn–Ce/TiO₂ catalyst, the decomposition temperature of NH₄HSO₄ was approximately 286 °C, which was much lower than 361 °C. This indicated that the thermal stability of NH₄HSO₄ on the catalyst was greatly reduced after introducing cerium. This inference was also proved by the DRIFT results. Therefore, ceria improved the performance of Mn/TiO₂ catalyst.

There is a universal agreement that residual SO₂ damages the metal oxide catalysts and decreases the NO conversion. (NH₄)₂SO₃ and (NH₄)₂SO₄ were formed on catalysts' surface by the reaction of SO₂. Researchers found that the NO conversion would increase for a while when SO₂ was introduced and then finally decrease. The adsorption of SO₂ improved the amount of Lewis acid sites, and thus the capacity of NH₃ was improved. However, the sulfation damages the manganese cations, which are the active sites of NO.

3.1.2 Sulfation of active sites. The presence of SO₂ could trigger the sulfation of the dominating active phase of metal oxide catalysts. Furthermore, the harm caused by the sulfation would be permanent and irreversible.¹²⁴ Jiang et al.¹²³ described a proposed mechanism of SO₂ deactivation effect for a Fe–Mn/Ti catalyst. The scheme in Fig. 6a shows the formation of Lewis acid sites. Mn cations are the active sites for the adsorption of NO to form bidentate or monodentate nitrates (Fig. 6b), however, when both NO and SO₂ exist in the flue gas, NO and SO₂ were adsorbed competitively. The adsorption ability of SO₂ was much higher than that of NO (Fig. 6c), so SO₂ occupied the active sites and the catalyst was sulfated. Furthermore, Fig. 6d shows that NH₃ could be adsorbed on the Lewis acid site of the Mn cations. When the active sites were sulfated, the Lewis acid sites could be transformed to the Brønsted acid sites via bonding of a water molecule. Therefore, this did not affect the adsorption of NH₃, because NH₃ could also be adsorbed on the Brønsted acid sites (Fig. 6e). It is therefore, proposed that the effect of SO₂ was mainly on the adsorption of NO rather than on the adsorption of NH₃.

Fig. 6 The proposed mechanism of SO₂ deactivation effect on the SCR reaction. (Reprinted with permission from ref. 123. Copyright (2010) American Chemical Society.)

Yu et al.¹¹¹ investigated the formation of metal sulfation on fresh Mn–Fe–Ce–Ti catalyst impregnated (NH₄)₂SO₄. In terms of the TG curve, SO₃ was released from (NH₄)₂SO₄ decomposition and then combined with Mn species to form manganese sulfate (MnSO₄). They claimed that the MnSO₄ could not be formed directly by the reaction of oxidized SO₂ and Mn species.⁵⁷ Kijlstra et al.¹²⁵ proved that the transformation of MnO to MnSO₄ on MnO_x/Al₂O₃ catalyst significantly deactivated the catalyst's activity.

Efforts have been made to facilitate the SO₂ tolerance of metal oxide catalysts. Ceria may trap SO₂ for NO_x storage catalysts to limit the sulfation of the dominating active phase and inhibit the formation of (NH₄)₂SO₄ and NH₄HSO₄.^117,126 After pre-treatment with SO₂, Ce doped Mn/TiO₂ catalysts had more Lewis acid sites than Mn/TiO₂ catalysts. This result implied that the addition of ceria could prevent the Lewis acid sites from the sulfation of SO₂.

Liu et al.¹²⁷ compared the performance of Mn–Ce mixed oxide catalysts prepared using the surfactant template method and the conventional co-precipitation method. Referring to the catalytic activity measurement, the Mn₅–Ce₅ catalyst prepared using the surfactant template method showed the highest NO_x conversion whether SO₂ and H₂O were introduced or not. The catalysts prepared using the surfactant template method possessed a higher surface area and smaller active sites, which contributed to a higher NO_x reduction.

In terms of in situ DRIFT analysis, Jin et al.²⁵ found that the Lewis acid sites could be preserved effectively with the doping of Ce while the SO₂ was added. SO₂ was oxidized to SO₃ or sulfation species on MnO_x, however, SO₃ and sulfation species move into ceria to form bulk like sulfate species. Therefore, ceria trapped SO₂ and protected the dominant active manganese cations (Fig. 7). Furthermore, in terms of the DRIFT and TG-DSC results, it was indicated that the thermal stability of sulfation species over the Mn–Ce catalyst was lower than that over the MnO_x catalyst. Referring to the study of Kylhammar et al.,¹²⁸ it is assumed that the bulk sulfation species in ceria reveals a high mobility, which facilitates their desorption.

Fig. 7 The formation schematic of bulk like sulfate on Mn–Ce/Ti catalysts. (Reprinted from ref. 25. Copyright 2013, with permission from Elsevier.)

Wang et al.,¹¹³ Xu et al.,¹²⁰ and Shi et al.¹²⁹ proved that the active manganese cation was reserved for ceria, which finally sulfated it. Ce⁴⁺ distributed on the catalysts' surface transformed into Ce³⁺ after sulfation. The reaction can be described as follows:

2CeO₂ + 3SO₂ + O₂ → Ce₂(SO₄)₃ (46)

Furthermore, it is reported that Zr could optimize the redox property and strengthen SO₂ tolerance.⁸⁹ Chang et al.^130,131 reported that Sn modification could further improve the tolerance of the Mn–Ce catalyst to SO₂ and H₂O. They compared the NO conversion of Sn(0.1)–Mn(0.4)–Ce(0.5)–O and Mn(0.4)–Ce(0.6)–O mixed oxide catalysts. It was obvious that the NO conversion of the Mn(0.4)–Ce(0.6)–O catalyst was decreased more significantly than that of the Sn(0.1)–Mn(0.4)–Ce(0.5)–O catalyst when 200 ppm of SO₂ and 3% O₂ was fed in to the system at 220 °C.

Shi et al.¹³² compared the resistance of the Mn/TiO₂ catalyst and the hierarchically macro-mesoporous Mn/TiO₂ (HM-Mn/TiO₂) catalyst prepared by the sol–gel method. After feeding 30 ppm SO₂ to the system, the NO conversion of the Mn/TiO₂ catalyst decreased sharply from 57% to 15%, however, the NO conversion of the HM-Mn/TiO₂ catalyst kept a higher value of more than 84%. The result indicated that maybe the SO₂ resistance could be improved by using a hierarchically macro-mesoporous structure.

As previously, because NH₃ could be adsorbed on both the Lewis acid sites and the Brønsted acid sites, there is little influence on the adsorption of NH₃. However, the adsorption ability of SO₂ was higher than that of NO. Residual SO₂ would be adsorbed on Mn cations, which are the active sites for the adsorption of NO. The damage caused by sulfation would be permanent and irreversible. Doping with ceria should be a good choice to divert this damage from Mn. More research should be done to investigate the reaction mechanism between SO₂ and Mn cations. The correlations should be established between the extent of sulfation and the degree of dispersion of MnO_x species at the surface.

3.1.3 Effect of H₂O. Water vapor could decrease the activity and show a notable inhibition on low temperature SCR. H₂O can be generated from the original flue gas or the reaction of SCR of NO. Even though there is no H₂O in the original flue gas, H₂O vapor will be generated during the SCR reaction, as shown in eqn (1). This means that the presence of H₂O is nearly inevitable. Therefore, many efforts have been made to evaluate the durability of metal oxide catalysts in the presence of H₂O vapor. As mentioned previously, trace SO₂ could still decrease the activity of the metal catalyst. The deactivation process of SO₂ would be enhanced in the case of H₂O vapor.

The main reason for the decrease of activity can be attributed to the competitive adsorption of H₂O. Many researchers reported that the adsorption of H₂O on the catalysts' surface blocked the active sites, which are provided for the adsorption of NH₃ and NO.^109,133 Chen et al.¹³⁴ studied a MnO_x–niobium oxides (NbO_x)–CeO₂ catalyst prepared by a sol–gel method and found that the adsorption of H₂O inhibited the adsorption of NO_x. Xiong et al.¹⁰⁸ compared the SCR performance of Mn–Fe spinel catalysts in the presence and absence of H₂O. They proposed that the effect of H₂O can be attributed to the competitive adsorption, the decrease of oxidation ability and the inhibition of interface reactions.^135,136 The temperature programmed desorption (TPD) profiles of NH₃ and NO_x were obtained, and the NO_x and NH₃ adsorption capacity of Mn–Fe spinel in the absence of H₂O and in the presence of 5% H₂O are shown in Table 3.

Table 3 Capacity of Mn–Fe spinel for NH₃ and NO_x adsorption at 150 °C μmol⁻¹ g⁻¹¹⁰⁸

Condition	NH3 (μmol^−1 g^−1)	NOx (μmol^−1 g^−1)
In the absence of H2O	122	82
In the presence of 5% H2O	105	46

---

Fig. 8 shows that the NO_x conversion apparently decreased when 5% H₂O was fed in to the flue gas, especially at the lower temperature, e.g., below 160 °C. The adsorption of H₂O vapor on the catalyst's active sites deprived the sites of NH₃ adsorption, which apparently decreased the NO conversion. There is a summary of Mn-containing catalysts' performance in the presence and in the absence of SO₂ and H₂O (Table 4).

Fig. 8 Dependence of NO conversion rate on gaseous NO concentration over Mn–Fe spinel: (a) in the absence of H₂O; (b) in the presence of 5% H₂O. (Reproduced from ref. 108 with permission from the Royal Society of Chemistry.)

Table 4 Summary of Mn-containing catalysts' performance in the absence/presence of SO₂ and H₂O

Catalysts	Preparation process^a	Reaction conditions^b	NOx conversion 1^c	Poison condition^d	NOx conversion 2^e	Ref.
a Preparation process means of preparation method, calcination temperature and time.b Reaction gas mixture and GHSV.c NO conversion at a specified temperature.d The concentration of SO2 and H2O introduced on the basis of reaction gas.e NO conversion at a certain temperature after introducing SO2 and/or H2O for a specified time.
Mn0.4–Ce0.5–Sn0.1–O	Co-precipitation/500 °C/6 h	0.1% NO, 0.1% NH3, 2% O2/35 000 h^−1	100% (110–230 °C)	0.01% SO2, 9% H2O	62% (110 °C)	130
Mn0.2–Ce0.1–Ti0.7–O	Hydrothermal/500 °C/6 h	0.05% NO, 0.05% NH3, 5% O2/64 000 h^−1	>92% (150–250 °C)	0.005% SO2, 5% H2O	∼90% (200 °C/10 h)	96
Mn0.3–Ce0.7–O	Citric acid/650 °C/6 h	0.1% NO, 0.1% NH3, 2% O2/42 000 h^−1	>95% (100–150 °C)	0.01% SO2, 2.5% H2O	∼95% (120 °C/4 h)	76
Mn5–Ce5–O	Surfactant template/500 °C/4 h	0.05% NO, 0.05% NH3, 5% O2/64 000 h^−1	>95% (100–200 °C)	0.005% SO2, 5% H2O	>90% (150–200 °C/-)	127
Mn0.28–Ce0.05–Ti 0.67–O	Co-precipitation/400 °C/2 h	0.06% NO, 0.06% NH3, 3% O2/40 000 h^−1	>92% (120–180 °C)	0.07% SO2, 3% H2O	35% (120 °C/13 h)	137
Mn0.4–Ce0.07–Ti1–O	Sol–gel/500 °C/6 h	0.08% NO, 0.08% NH3, 3% O2/40 000 h^−1	∼100% (100–180 °C)	0.01% SO2, 3% H2O	∼60% (100 °C/10 h)	138
Sn0.1–Mn0.4–Ce0.5–O	Co-precipitation/500 °C/6 h	0.1% NO, 0.1% NH3, 2% O2/35 000 cm^3 g^−1 h^−1	∼100% (110–230 °C)	0.01% SO2, 12% H2O	∼70% (110 °C/9 h)	131
Mn1–Ce0.3/TiO2–graphene	Impregnation/500 °C/6 h	0.05% NO, 0.05% NH3, 7% O2/67 000 h^−1	>90% (140–180 °C)	0.02% SO2, 10% H2O	∼75% (180 °C/3 h)	117
Mn0.6/Ce0.5–Zr0.5–O	Impregnation/500 °C/6 h	0.06% NO, 0.06% NH3, 3% O2/30 000 h^−1	>90% (140–180 °C)	0.01% SO2, 3% H2O	∼90% (180 °C/3 h)	139
Mn0.4–Ce0.07–Ti1–O	Sol–gel/500 °C/6 h	0.1% NO, 0.1% NH3, 3% O2/40 000 h^−1	∼100% (120–220 °C)	0.01% SO2, 3% H2O	∼82% (150 °C/7 h)	140
Mn0.4–Ce0.5–W0.1–O	Sol–gel/600 °C/3 h	0.05% NO, 0.05% NH3, 5% O2/40 000 h^−1	>80% (140–300 °C)	0.006% SO2, 5% H2O	∼55% (150 °C/3 h)	141
Mn0.3–Ce0.7–O	Citric acid/650 °C/6 h	0.1% NO, 0.1% NH3, 2% O2/42 000 h^−1	∼100% (120–150 °C)	0.01% SO2, 6% H2O	∼92% (120 °C/4 h)	23
Mn0.2–Ce0.1–Ti0.7–O	Hydrothermal/500 °C/6 h	0.05% NO, 0.05% NH3, 5% O2/64 000 h^−1	>95% (150–350 °C)	0.005% SO2, 5% H2O	∼90% (200 °C/10 h)	96
Mn–Ce–W–Ti–O	Impregnation/400 °C/4 h	0.02% NO, 0.02% NH3, 8% O2/30 000 h^−1	∼100% (160–200 °C)	0.01% SO2, 8% H2O	∼85% (180 °C/10 h)	97
Mn0.4–Ce0.07–Ti1–O	Co-precipitation/400 °C/2 h	0.06% NO, 0.06% NH3, 3% O2/40 000 h^−1	>92% (120–180 °C)	0.07% SO2, 3% H2O	61% (120 °C/2.5 h)	142
Mn0.2–Fe0.15–Ce0.3–Ti1–O	Sol–gel/500 °C/6 h	0.06% NO, 0.06% NH3, 3% O2/50 000 h^−1	>95% (160–260 °C)	0.01% SO2, 3% H2O	∼85% (180 °C/6 h)	143
Mn0.6–Ce0.5–Zr0.5–O	Impregnation/500 °C/6 h	0.06% NO, 0.066% NH3, 6% O2/45 000 h^−1	>95% (140–220 °C)	0.01% SO2, 3% H2O	∼90% (180 °C/7 h)	144
Mn0.4–Fe0.1–Ce0.5–O	Co-precipitation/500 °C/6 h	0.1% NO, 0.1% NH3, 2% O2/84 000 h^−1	>82% (150–180 °C)	0.01% SO2, 2.5% H2O	>90% (150 °C/4 h)	27
Mn0.4–Ce0.1–Ti1–O	Sol–gel/500 °C/6 h	0.08% NO, 0.08% NH3, 3% O2/40 000 h^−1	∼100% (150 °C)	0.01% SO2, 3% H2O	∼60% (150 °C/10 h)	25
Mn0.6–Fe0.4–O	Citric acid/500 °C/3 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	>95% (90–220 °C)	0.01% SO2, 5% H2O	∼88% (120 °C/6 h)	98
10% Mn/Fe–Ti spinel	Impregnation/500 °C/3 h	0.05% NO, 0.05% NH3, 2% O2/24 000 cm^3 g^−1 h^−1	>95% (150–250 °C)	0.006% SO2, 8% H2O	∼80% (200 °C/13 h)	53
Mn0.4–Fe0.1/Ti1–Zr0.5–O	Sol–gel/500 °C/6 h	0.1% NO, 0.1% NH3, 4% O2/30 000 h^−1	>95% (80–180 °C)	0.01% SO2, 8% H2O	∼70% (150 °C/5 h)	89
Mn0.4–Fe0.1/Ti0.5–O	Sol–gel/500 °C/6 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	∼100% (150 °C)	0.02% SO2	∼65% (150 °C/6 h)	123
Mn0.6–Ti1–O	Sol–gel/500 °C/6 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	∼100% (110 °C)	0.003% SO2	∼85% (120 °C/9 h)	132
Mn0.4–Ti1–O	Sol–gel/500 °C/6 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	>90% (180–250 °C)	0.02% SO2, 3% H2O	∼70% (150 °C/6 h)	118
Mn–Fe–Ce–Ti–O	Sol–gel/500 °C/6 h	0.06% NO, 0.048% NH3, 2% O2/24 000 h^−1	>80% (200–300 °C)	0.03% SO2, 10% H2O	∼80% (240 °C/5 h)	111
7% Mn/Ti1–graphene	Impregnation/450° C/6 h	0.05% NO, 0.05% NH3, 7% O2/67 000 h^−1	>80% (120–180 °C)	0.02% SO2, 10% H2O	∼72% (180 °C/3 h)	115
Mn0.5–Zr0.5–O	Citric acid/450 °C/3 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	∼100% (100–200 °C)	0.01% SO2, 5% H2O	∼40% (150 °C/13 h)	124
Mn2.5–Cu0.1–Ti1–O	Co-precipitation/350 °C/6 h	0.05% NO, 0.05% NH3, 5% O2/30 000 h^−1	∼100% (60–280 °C)	0.01% SO2, 11% H2O	∼60% (125 °C/10 h)	39
Mn4/Co0.6–Ce2.7–Zr2.7	Impregnation/500 °C/6 h	0.06% NO, 0.06% NH3, 6% O2/45 000 h^−1	>97% (120–220 °C)	0.01% SO2, 3% H2O	∼93% (180 °C/7 h)	109
Mn0.6–Cr0.4–O	Citric acid/650 °C/3 h	0.1% NO, 0.1% NH3, 3% O2/30 000 h^−1	∼100% (120–220 °C)	0.01% SO2	∼82% (120 °C/5 h)	37
Mn0.3–Ce0.7–O	Citric acid/650 °C/6 h	0.05% NO, 0.05% NH3, 2% O2/30 000 cm^3 g^−1 h^−1	>95% (120–160 °C)	5% H2O	∼70% (120–140 °C)	61
Mn0.3–Ce0.7–O	Citric acid/650 °C/6 h	0.05% NO, 0.05% NH3, 2% O2/120 000 cm^3 g^−1 h^−1	>90% (140–200 °C)	5% H2O	>70% (160–200 °C)	88
Mn–Ce/W–Zr–O	Impregnation/550 °C/3 h	0.1% NO, 0.1% NH3, 5% O2/10 000 h^−1	>90% (150–250 °C)	0.01% SO2, 10% H2O	<80% (140–240 °C)	145
Mn–Ce/activated carbon honeycomb	Impregnation/400 °C/3 h	0.05% NO, 0.05% NH3, 5% O2/1910 h^−1	∼84% (160 °C)	0.03% SO2	∼44% (160 °C/7 h)	113
Mn0.23–Nb0.23–Ce0.54–O	Co-precipitation/650 °C/5 h	0.1% NO, 0.1% NH3, 10% O2/52 000 h^−1	—	5% H2O	>80% (200–300 °C)	146
Mn0.23–Nb0.23–Ce0.54–O	Co-precipitation/650 °C/5 h	0.1% NO, 0.1% NH3, 10% O2/52 000 h^−1	>78% (200–300 °C)	0.005% SO2, 5% H2O	∼20% (250 °C/0.5 h)	147
Mn2.5–La2.5–Ce1–Ni1	500 °C/6 h	0.06% NO, 0.06% NH3, 6% O2/20 000 h^−1	∼98% (150–350 °C)	0.03% SO2	∼85% (200 °C/4 h)	148

---

3.1.4 Regeneration. Many articles reported that the deactivated (NH₄)₂SO₄ could be regenerated after use. Water washing, thermal regeneration, thermal reduction regeneration and reductive regeneration were the usual methods to regenerate the deactivated catalysts.^137,149 Yu et al.¹⁵⁰ investigated the regeneration of the SCR catalyst using dilute sodium hydroxide solution. The catalyst was deactivated by the deposition of sulfates on the surface. Pourkhalil et al.¹⁵¹ regenerated the deactivated MnO_x catalysts via heating at 350 °C for 2 h. This was a reversible process because of (NH₄)₂SO₄ salts can be decomposed. Jin et al.²⁵ regenerated the Mn/Ti and Mn–Ce/Ti catalysts with water washing (Fig. 9a). Shi et al.¹²⁹ regenerated the CeO₂ catalysts using a thermal treatment (Fig. 9b).

Fig. 9 SCR activities of Mn/Ti and Mn–Ce/Ti in the presence of SO₂. (a) Reaction conditions: [NO] = [NH₃] = 800 ppm, [O₂] = 3%, [SO₂] = 100 ppm, [H₂O] = 3 vol%, balance N₂, temperature: 150 °C, GHSV = 40 000 h⁻¹. (Reprinted from ref. 25. Copyright 2013, with permission from Elsevier.) (b) Regeneration of sulfur poisoned CeO₂ catalyst using a thermal treatment. Reaction conditions: [NO] = [NH₃] = 500 ppm, [O₂] = 5%, [SO₂] = 25 ppm, balance N₂, temperature: 350 °C, GHSV = 175 000 h⁻¹. (Reprinted from ref. 129. Copyright 2016, with permission from Elsevier.)

Huang et al.¹⁵² investigated a series of Fe–Mn oxide catalysts supported on mesoporous silica (MPS), which showed good activity. When H₂O and SO₂ was fed in to the system at 190 °C, the NO conversion over Mn–Fe/MPS was finally decreased to 85.3% from 99.2%. This was attributed to the formation of the NH₄HSO₄ and (NH₄)₂SO₄ in the presence of both H₂O and SO₂. However, the deactivated catalyst could be regenerated using a heating treatment, because the deactivation was because of the catalyst pore plugging and surface area loss by the deposition of (NH₄)₂SO₄. When the temperature is above 140 °C, H₂O has no negative effect on its activity.

Guan et al.¹⁵³ investigated the resistance to deactivation by H₂O and SO₂ of Ti_0.9Ce_0.05V_0.05O_2−x catalysts, which showed a high NO conversion and N₂ selectivity. After feeding 400 ppm SO₂ for 26 h at 150 °C, the surface of catalyst was deposited with significant agglomeration and bulk NH₄NO₃ and (NH₄)₂SO₄ with a size of 30–50 μm. Then, the NH₄NO₃ and (NH₄)₂SO₄ was decomposed when the catalyst was calcined at 200 °C and 400 °C, because the decomposition temperatures were 170 °C and 300 °C, respectively. The surfaces were scanned using scanning electron microscopy (SEM), and the transformation of the surface is shown in Fig. 10.

Fig. 10 SEM images (a) feeding with 400 ppm SO₂ at 150 °C for 26 h, (b) calcined at 200 °C and 400 °C for 2 h. (Reprinted from ref. 153. Copyright 2011, with permission from Elsevier.)

3.2 Alkali and alkaline earth metal ions

Fine fly ash still exists in the downstream of the flue gas after desulfurizing and dedusting. Amounts of alkali and alkaline earth metals were released from the raw materials or coal, such as in the cement production process. Alkali salts are important components in fine fly ash, which not only plugs the pores of catalysts, but also decreases SCR activity by reacting with the active phase.^154–156 In addition, because of the water solubility or ion exchange, alkali metal has a high liquidity to neutralize the acid sites.¹⁵⁷ For the traditional V₂O₅-based SCR catalysts, alkali metal deactivated these by affecting the acid sites on the surface.^154,158 Alkali metals could lower MnO_x reducibility, decrease specific surface areas and damage the acid sites of low temperature catalysts.

Zhou et al.¹⁵⁹ reported that sodium sulfate, used to simulate the combined effects of alkali metal and SO₂ in the flue gas, had strong effects on the activity of the Mn–Ce/TiO₂ catalyst, such as simultaneous pore occlusion and sulfation effect. Guo et al.¹⁶⁰ investigated the deactivation effect of sodium (Na) and potassium (K) on a Mn/TiO₂ catalyst. The catalyst was prepared using a sol–gel method and Na and K were doped via an impregnation method. The Mn/TiO₂ catalysts exhibited a high activity of 90% NO conversion. However, when Na or K was doped, the conversion was decreased from 95% to 78% and 27%, respectively. In this study, the effect of K was apparently more serious than that of Na.¹⁶¹ Furthermore, Chen et al.¹⁵⁵ found that on the catalysts' surface chemisorbed oxygen was reduced by alkali and alkaline earth ions together with a decrease of SCR activity. The downward trend was K > Na > Ca > Mg.

Shen et al.^162,163 studied the effects of K, Na and Ca on a Mn–Ce/Zr catalyst. From the NH₃-TPD measurements, the adsorption of NH₃ was decreased when the catalyst was doped with alkali metal ions. This may indicate that the alkali metal on the surface of the catalysts may destroy the surface acidic sites, and decrease the redox property and chemisorbed oxygen. Furthermore, they also found that K was more harmful to the catalyst compared to Na or Ca. However, Kustov et al.¹⁶⁴ found that V₂O₅ supported on sulfated zirconium dioxide showed a good resistance towards alkali ions. Chen et al.⁷⁰ reported that the K resistance of the Mn/TiO₂ catalyst could be improved by doping it with Co, which increased the adsorption of NH₃ and NO_x species.

3.3 Heavy metal ions

Heavy metal ions, regarded as hazardous pollutants, can deactivate the SCR catalysts. Heavy metal ions in the flue gas are mainly generated from coal used as fuel.¹⁶⁵ It has been proved that heavy metals could lead to the deactivation of vanadium-based SCR catalysts.¹⁶⁶ Kong et al.¹⁶⁷ found that the Brønsted acid sites of a V–W/TiO₂ catalyst were impacted when mercury chloride was introduced. Actually, there is little heavy metal ions found in the downstream of the precipitator because the heavy metal ions usually exist in the fly ash. Moreover, water vapor exists in the flue gas all along. For the water solubility of heavy metal ions, it is necessary to take the effect of heavy metals into consideration.

Lead (Pb) and zinc (Zn) are typical heavy metals found in the flue gas of coal fired power plants. Guo et al.^156,168, and Li et al.¹⁶⁹ compared the poisoning effect of Pb and Zn on a Mn/TiO₂ catalyst. The Pb or Zn was loaded on to the Mn/TiO₂ catalyst using impregnation. As a result, both Pb and Zn were found to have a negative effect on the Mn/TiO₂ catalyst (Fig. 11a). From the characterization experiments, the redox ability of Zn–Mn/TiO₂ and Pb–Mn/TiO₂ was found to be decreased because of the drop of Mn⁴⁺ and chemisorbed oxygen. Zhou et al.¹⁷⁰ investigated the deactivation effects of lead(II) oxide (PbO) on the Mn–Ce/TiO₂ catalyst. It was proposed that the surface area, the concentration of Mn⁴⁺, Ce³⁺ and chemisorbed oxygen was decreased after introducing PbO. Consequently, the performance of the Mn–Ce/TiO₂ catalyst was greatly decreased because of the poisoning of PbO (Fig. 11b).

Fig. 11 NO conversion over pure and poisoned catalysts. (a) Reaction conditions: [NO] = [NH₃] = 600 ppm, [O₂] = 5%, balance N₂, GHSV = 108 000 h⁻¹. (Reprinted from ref. 156. Copyright 2015, with permission from Elsevier.) (b) Reaction conditions: [NO] = [NH₃] = 800 ppm, [O₂] = 5%, balance N₂, GHSV = 200 000 h⁻¹. (Reprinted from ref. 170. Copyright 2016, with permission from Elsevier.)

Mercury (Hg⁰) is a toxic trace element in the atmosphere and has a high concentration in coals used in China, such as anthracite, bituminous coal and lignite.¹⁷¹ Researchers have attempted to remove the NO and Hg⁰ simultaneously. However, Hg⁰ is harmful to the catalysts of SCR of NO because it will compete with NH₃ for adsorption on the active sites.¹⁷² Xu et al.⁴¹ investigated the influence of Hg⁰ on the NO conversion over a LaMnO₃ catalyst. The NO conversion had a slight decrease in the presence of Hg⁰ (Fig. 12).

Fig. 12 The effect of Hg⁰ on NO conversion. Reaction conditions: [NO] = [NH₃] = 500 ppm, [Hg⁰] = 500 μg m⁻³, [O₂] = 4%, balance N₂, GHSV = 478 000 h⁻¹. (Reprinted from ref. 41. Copyright 2016, with permission from Elsevier.)

4. Conclusions and perspectives

NH₃-SCR of NO_x in the presence of O₂ is one of the important strategies in controlling NO_x emissions. Low temperature SCR has been investigated for several decades. Mn-containing metal oxide catalysts generally gave the preferable performance. SCR of NO_x with NH₃ follows both the L–H and the E–R mechanisms. There is quite a similarity between these two different mechanisms. A comproportionation occurs in both the L–H and E–R mechanisms. Fast SCR has a higher reaction rate than standard SCR and it depends on the formation of NO₂. N₂O formation can mainly be explained using the E–R mechanism. A synergistic mechanism is vital for designing a remarkable metal oxide catalyst. Multi-metal cations will promote the performance mutually. Manganese cations mainly serve as the adsorption center for nitrogen. Thus, it is necessary to introduce an element for the adsorption of oxygen and to provide a redox cycle.

A big challenge in the industrial use of Mn-containing oxide catalysts is their durability. They are vulnerable to the effects of both SO₂ and H₂O. Sulfur oxides and water vapor cause the deactivation of Mn-containing catalysts. Alkali metals could lower manganese oxide reducibility, decrease specific surface areas and damage the acid sites of low temperature catalysts. The poisoning process of SO₂ can be classified into two categories: deposition of (NH₄)₂SO₄ and sulfation of the active phase. For the low temperature downstream of the flue gas, the deposition of (NH₄)₂SO₄ or NH₄HSO₄ occurs more easily and NH₃ is evidently adsorbed by H₂O in comparison with the operation upstream. Many efforts have been made to improve the durability. Nonetheless, few techniques have been useful in practical industrial applications.

On the basis of the previous analysis, some conclusions can be drawn as follows:

(1) Most research is related to the performance of the catalysts, such as NO conversion, N₂ selectivity and poisons' tolerance, as well as the mechanism of this process. An excellent NO conversion of catalysts has been obtained, however, the N₂ selectivity is not satisfactory.

(2) Less effort has been made on determining the relationship of metal oxide crystal structure and its performance, which is required for the design of catalysts. More attention should be given to the relationship between the catalysts' structure and its reaction mechanism, which guides us exactly to design a low temperature SCR catalyst for different flue gases.

(3) Mn-containing metal oxide catalysts show a notable SCR performance at low temperature. However, the single manganese oxide catalysts have a poor tolerance of SO₂ and H₂O, which has been improved by modifying other elements in bench scale experiments. Researchers have been engaged in improving Mn-containing catalysts by modifying them with different metal oxides. Ce can enhance the adsorption of NO and O₂ which benefits the oxidization of NO to NO₂ and improves sulfur resistance, and inhibits the formation of (NH₄)₂SO₄ and NH₄HSO₄. Ce has good selectivity for improving the catalysts' performance. More research efforts should be made on the activity and poisoning tolerance.

(4) Most catalysts were powder rather than monolith catalysts, such as honeycomb or slab. A laboratory study is a small scale test that will react differently to industrial tests. Specific surface area is important to the activity and closely related to the particles' size, shape and aggregation. The preparation method is also important to the catalysts' performance. Researchers should give more attention to pilot scale tests or industrial tests.

(5) The low temperature SCR catalysts have been investigated for several decades. Lots of elements have been studied in the catalysts. To avoid repetitive work and waste of resources, a low temperature SCR catalysts' materials database should be built.

(6) Heaps of disabled SCR catalysts should be regenerated and reused. The regeneration and recycling of SCR catalysts is another big task for researchers. This problem should be taken into consideration while researchers are designing new SCR catalysts.

Acknowledgements

This work was sponsored by the National Natural Science Foundation of China (Grants U1360202, 51472030, 51672024 and 51502014) and the 111 Project (No. B17003). The authors would like to thank the editor for editing the manuscript and the anonymous reviewers for their detailed and helpful comments.

References

1. H. Amini, S.-M. Taghavi-Shahri, S. B. Henderson, V. Hosseini, H. Hassankhany, M. Naderi, S. Ahadi, C. Schindler, N. Künzli and M. Yunesian, Sci. Rep., 2016, 6, 32970.
2. A. Richter, J. P. Burrows, H. Nusz, C. Granier and U. Niemeier, Nature, 2005, 437, 129–132.
3. J. N. Galloway, F. J. Dentener, D. G. Capone, E. W. Boyer, R. W. Howarth, S. P. Seitzinger, G. P. Asner, C. C. Cleveland, P. A. Green, E. A. Holland, D. M. Karl, A. F. Michaels, J. H. Porter, A. R. Townsend and C. J. Vöosmarty, Biogeochemistry, 2004, 70, 153–226.
4. J. Zhu and A. Thomas, Appl. Catal., B, 2009, 92, 225–233.
5. M. Fu, C. Li, P. Lu, L. Qu, M. Zhang, Y. Zhou, M. Yu and Y. Fang, Catal. Sci. Technol., 2014, 4, 14–25.
6. C. Tang, H. Zhang and L. Dong, Catal. Sci. Technol., 2016, 6, 1248–1264.
7. B. Liu, Y. H. Wang and H. Xu, J. Energy Eng., 2016, 142, 10.
8. B. Liu, B. Bao, Y. Wang and H. Xu, J. Energy Inst., 2016, 90, 441–451.
9. T. Ishihara, J. Catal., 2003, 220, 104–114.
10. D. G. Streets and S. T. Waldhoff, Atmos. Environ., 2000, 34, 363–374.
11. P. Forzatti, Appl. Catal., A, 2001, 222, 221–236.
12. M. Koebel, G. Madia and M. Elsener, Catal. Today, 2002, 73, 239–247.
13. M. T. Javed, N. Irfan and B. M. Gibbs, J. Environ. Manage., 2007, 83, 251–289.
14. S. W. Bae, S. A. Roh and S. D. Kim, Chemosphere, 2006, 65, 170–175.
15. J. O. L. Wendt, W. P. Linak, P. W. Groff and R. K. Srivastava, AIChE J., 2001, 47, 2603–2617.
16. S. Roy and A. Baiker, Chem. Rev., 2009, 109, 4054–4091.
17. G. Busca, L. Lietti, G. Ramis and F. Berti, Appl. Catal., B, 1998, 18, 1–36.
18. K.-I. Shimizu, J. Shibata, H. Yoshida, A. Satsuma and T. Hattori, Appl. Catal., B, 2001, 30, 151–162.
19. S. Brandenberger, O. Kroecher, A. Tissler and R. Althoff, Catal. Rev.: Sci. Eng., 2008, 50, 492–531.
20. L. Casagrande, L. Lietti, I. Nova, P. Forzatti and A. Baiker, Appl. Catal., B, 1999, 22, 63–67.
21. K. Bourikas, C. Fountzoula and C. Kordulis, Appl. Catal., B, 2004, 52, 145–153.
22. S. Hodjati, K. Vaezzadeh, C. Petit, V. Pitchon and A. Kiennemann, Top. Catal., 2001, 16, 151–155.
23. G. Qi and R. T. Yang, Chem. Commun., 2003, 848–849,  10.1039/b212725c.
24. R. Qu, X. Gao, K. Cen and J. Li, Appl. Catal., B, 2013, 142–143, 290–297.
25. R. Jin, Y. Liu, Y. Wang, W. Cen, Z. Wu, H. Wang and X. Weng, Appl. Catal., B, 2014, 148–149, 582–588.
26. D. A. Peña, B. S. Uphade and P. G. Smirniotis, J. Catal., 2004, 221, 421–431.
27. G. Qi, R. T. Yang and R. Chang, Appl. Catal., B, 2004, 51, 93–106.
28. B. Thirupathi and P. G. Smirniotis, Appl. Catal., B, 2011, 110, 195–206.
29. W. Shan, F. Liu, H. He, X. Shi and C. Zhang, Appl. Catal., B, 2012, 115, 100–106.
30. 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.
31. F. Kapteijn, L. Singoredjo, A. Andreini and J. A. Moulijn, Appl. Catal., B, 1994, 3, 173–189.
32. F. ç. Larachi, J. Pierre, A. Adnot and A. Bernis, Appl. Surf. Sci., 2002, 195, 236–250.
33. B. Shen, F. Wang and T. Liu, Powder Technol., 2014, 253, 152–157.
34. F. Liu, H. He, C. Zhang, Z. Feng, L. Zheng, Y. Xie and T. Hu, Appl. Catal., B, 2010, 96, 408–420.
35. S. Yang, S. Xiong, Y. Liao, X. Xiao, F. Qi, Y. Peng, Y. Fu, W. Shan and J. Li, Environ. Sci. Technol., 2014, 48, 10354–10362.
36. H. Hu, S. Cai, H. Li, L. Huang, L. Shi and D. Zhang, ACS Catal., 2015, 5, 6069–6077.
37. Z. Chen, Q. Yang, H. Li, X. Li, L. Wang and S. Chi Tsang, J. Catal., 2010, 276, 56–65.
38. M. A. Zamudio, N. Russo and D. Fino, Ind. Eng. Chem. Res., 2011, 50, 6668–6672.
39. M. Kang, E. D. Park, J. M. Kim and J. E. Yie, Catal. Today, 2006, 111, 236–241.
40. D. Fang, J. Xie, D. Mei, Y. Zhang, F. He, X. Liu and Y. Li, RSC Adv., 2014, 4, 25540.
41. H. Xu, Z. Qu, C. Zong, F. Quan, J. Mei and N. Yan, Appl. Catal., B, 2016, 186, 30–40.
42. Z. Liu and S. Ihl Woo, Catal. Rev., 2006, 48, 43–89.
43. J. Li, H. Chang, L. Ma, J. Hao and R. T. Yang, Catal. Today, 2011, 175, 147–156.
44. C. Liu, J.-W. Shi, C. Gao and C. Niu, Appl. Catal., A, 2016, 522, 54–69.
45. http://www.sdpc.gov.cn/gzdt/<?ccdc_no 201409?><?ccdc_no 201409?>201409<?ccdc END?><?ccdc END?>/t20140919_626240.html, accessed Dec. 7th, 2016.
46. P. R. Ettireddy, N. Ettireddy, T. Boningari, R. Pardemann and P. G. Smirniotis, J. Catal., 2012, 292, 53–63.
47. G. Busca, M. A. Larrubia, L. Arrighi and G. Ramis, Catal. Today, 2005, 107–108, 139–148.
48. G. Zhou, B. Zhong, W. Wang, X. Guan, B. Huang, D. Ye and H. Wu, Catal. Today, 2011, 175, 157–163.
49. L. Qiu, J. Meng, D. Pang, C. Zhang and F. Ouyang, Catal. Lett., 2015, 145, 1500–1509.
50. D. Ye, Practical Inorganic Thermodynamic Data Manual, Metallurgical Industry Press, Beijing, 1st edn, 1981.
51. I. Nam, J. Catal., 1989, 119, 269.
52. S. Yang, F. Qi, Y. Liao, S. Xiong, Y. Lan, Y. Fu, W. Shan and J. Li, Ind. Eng. Chem. Res., 2014, 53, 5810–5819.
53. S. Yang, F. Qi, S. Xiong, H. Dang, Y. Liao, P. K. Wong and J. Li, Appl. Catal., B, 2016, 181, 570–580.
54. T. Chen, B. Guan, H. Lin and L. Zhu, Chin. J. Catal., 2014, 35, 294–301.
55. S. Yang, C. Wang, J. Li, N. Yan, L. Ma and H. Chang, Appl. Catal., B, 2011, 110, 71–80.
56. G. Qi and R. T. Yang, J. Phys. Chem. B, 2004, 108, 15738–15747.
57. F. Liu and H. He, Catal. Today, 2010, 153, 70–76.
58. Z. Wu, B. Jiang, Y. Liu, H. Wang and R. Jin, Environ. Sci. Technol., 2007, 41, 5812–5817.
59. D. Fang, F. He, D. Li and J. Xie, Appl. Surf. Sci., 2013, 285, 215–219.
60. D. Fang, J. Xie, H. Hu, H. Yang, F. He and Z. Fu, Chem. Eng. J., 2015, 271, 23–30.
61. S. Xiong, Y. Liao, X. Xiao, H. Dang and S. Yang, J. Phys. Chem. C, 2015, 119, 4180–4187.
62. F. Eigenmann, M. Maciejewski and A. Baiker, Appl. Catal., B, 2006, 62, 311–318.
63. X. Tang, J. Li, L. Sun and J. Hao, Appl. Catal., B, 2010, 99, 156–162.
64. I. Nova, C. Ciardelli, E. Tronconi, D. Chatterjee and B. Bandl-Konrad, Catal. Today, 2006, 114, 3–12.
65. M. Koebel, G. Madia, F. Raimondi and A. Wokaun, J. Catal., 2002, 209, 159–165.
66. M. Koebel, M. Elsener and G. Madia, Ind. Eng. Chem. Res., 2001, 40, 52–59.
67. G. Madia, M. Koebel, M. Elsener and A. Wokaun, Ind. Eng. Chem. Res., 2002, 41, 3512–3517.
68. M. Stanciulescu, P. Bulsink, G. Caravaggio, L. Nossova and R. Burich, Appl. Surf. Sci., 2014, 300, 201–207.
69. W. Zhao, C. Li, P. Lu, Q. Wen, Y. Zhao, X. Zhang, C. Fan and S. Tao, Environ. Technol., 2013, 34, 81–90.
70. Q.-L. Chen, R.-T. Guo, Q.-S. Wang, W.-G. Pan, N.-Z. Yang, C.-Z. Lu and S.-X. Wang, J. Taiwan Inst. Chem. Eng., 2016, 64, 116–123.
71. X. Zeng, X. Huo, T. Zhu, X. Hong and Y. Sun, Molecules, 2016, 21, 1491.
72. Y. Liang, Y. Huang, H. Zhang, L. Lan, M. Zhao, M. Gong, Y. Chen and J. Wang, Environ. Sci. Pollut. Res. Int., 2017, 24, 9314–9324.
73. A. Grossale, I. Nova and E. Tronconi, J. Catal., 2009, 265, 141–147.
74. L. Zhu, Z. Zhong, H. Yang and C. Wang, Water, Air, Soil Pollut., 2016, 227, 476.
75. A. Grossale, I. Nova, E. Tronconi, D. Chatterjee and M. Weibel, J. Catal., 2008, 256, 312–322.
76. G. Qi and R. T. Yang, J. Catal., 2003, 217, 434–441.
77. R. Q. Long, R. T. Yang and R. Chang, Chem. Commun., 2002, 452–453,  10.1039/b111382h.
78. L. J. France, Q. Yang, W. Li, Z. Chen, J. Guang, D. Guo, L. Wang and X. Li, Appl. Catal., B, 2017, 206, 203–215.
79. Y. Zhu, Y. Zhang, R. Xiao, T. Huang and K. Shen, Catal. Commun., 2017, 88, 64–67.
80. S. M. Mousavi, A. Niaei, M. J. Illán Gómez, D. Salari, P. Nakhostin Panahi and V. Abaladejo-Fuentes, Mater. Chem. Phys., 2014, 143, 921–928.
81. T. Wang, K. Sun, Z. Lu and Y. Zhang, React. Kinet., Mech. Catal., 2010, 101, 153–161.
82. Y. Liu, T. Gu, X. Weng, Y. Wang, Z. Wu and H. Wang, J. Phys. Chem. C, 2012, 116, 16582–16592.
83. S. Xiong, Y. Liao, H. Dang, F. Qi and S. Yang, RSC Adv., 2015, 5, 27785–27793.
84. L. Xu, X.-S. Li, M. Crocker, Z.-S. Zhang, A.-M. Zhu and C. Shi, J. Mol. Catal. A: Chem., 2013, 378, 82–90.
85. B. Jiang, Z. Li and S.-C. Lee, Chem. Eng. J., 2013, 225, 52–58.
86. S. Suarez, J. Martin, M. Yates, P. Avila and J. Blanco, J. Catal., 2005, 229, 227–236.
87. M. A. Zamudio, N. Russo and D. Fino, Ind. Eng. Chem. Res., 2011, 50, 6668–6672.
88. S. Yang, Y. Liao, S. Xiong, F. Qi, H. Dang, X. Xiao and J. Li, J. Phys. Chem. C, 2014, 118, 21500–21508.
89. B. Jiang, B. Deng, Z. Zhang, Z. Wu, X. Tang, S. Yao and H. Lu, J. Phys. Chem. C, 2014, 118, 14866–14875.
90. T.-Y. Li, S.-J. Chiang, B.-J. Liaw and Y.-Z. Chen, Appl. Catal., B, 2011, 103, 143–148.
91. R. Farra, M. García-Melchor, M. Eichelbaum, M. Hashagen, W. Frandsen, J. Allan, F. Girgsdies, L. Szentmiklósi, N. López and D. Teschner, ACS Catal., 2013, 3, 2256–2268.
92. S. Imamura, A. Doi and S. Ishida, Ind. Eng. Chem. Prod. Res. Dev., 1985, 24, 75–80.
93. D. A. Peña, B. S. Uphade, E. P. Reddy and P. G. Smirniotis, J. Phys. Chem. B, 2004, 108, 9927–9936.
94. H. Chang, J. Li, J. Yuan, L. Chen, Y. Dai, H. Arandiyan, J. Xu and J. Hao, Catal. Today, 2013, 201, 139–144.
95. R. Jin, Y. Liu, Z. Wu, H. Wang and T. Gu, Chemosphere, 2010, 78, 1160–1166.
96. Z. Liu, J. Zhu, J. Li, L. Ma and S. I. Woo, ACS Appl. Mater. Interfaces, 2014, 6, 14500–14508.
97. D. W. Kwon, K. B. Nam and S. C. Hong, Appl. Catal., A, 2015, 497, 160–166.
98. Z. Chen, F. Wang, H. Li, Q. Yang, L. Wang and X. Li, Ind. Eng. Chem. Res., 2012, 51, 202–212.
99. Y. J. Kim, H. J. Kwon, I.-S. Nam, J. W. Choung, J. K. Kil, H.-J. Kim, M.-S. Cha and G. K. Yeo, Catal. Today, 2010, 151, 244–250.
100. Z. Liu, Y. Liu, Y. Li, H. Su and L. Ma, Chem. Eng. J., 2016, 283, 1044–1050.
101. Z. Liu, Y. Li, T. Zhu, H. Su and J. Zhu, Ind. Eng. Chem. Res., 2014, 53, 12964–12970.
102. P. Lu, C. Li, G. Zeng, L. He, D. Peng, H. Cui, S. Li and Y. Zhai, Appl. Catal., B, 2010, 96, 157–161.
103. Y. Shen, S. Zhu, T. Qiu and S. Shen, Catal. Commun., 2009, 11, 20–23.
104. G. Ramis, J. Catal., 1990, 124, 574–576.
105. G. Busca, Appl. Catal., B, 1998, 18, 1–36.
106. P. R. Ettireddy, N. Ettireddy, S. Mamedov, P. Boolchand and P. G. Smirniotis, Appl. Catal., B, 2007, 76, 123–134.
107. M. Inomata, J. Catal., 1980, 62, 140–148.
108. S. Xiong, Y. Liao, X. Xiao, H. Dang and S. Yang, Catal. Sci. Technol., 2015, 5, 2132–2140.
109. X. Zhang, B. Shen, K. Wang and J. Chen, J. Ind. Eng. Chem., 2013, 19, 1272–1279.
110. X. Wang, X. Li, Q. Zhao, W. Sun, M. Tade and S. Liu, Chem. Eng. J., 2016, 288, 216–222.
111. J. Yu, F. Guo, Y. Wang, J. Zhu, Y. Liu, F. Su, S. Gao and G. Xu, Appl. Catal., B, 2010, 95, 160–168.
112. G. Xie, Z. Liu, Z. Zhu, Q. Liu, J. Ge and Z. Huang, J. Catal., 2004, 224, 36–41.
113. Y. Wang, X. Li, L. Zhan, C. Li, W. Qiao and L. Ling, Ind. Eng. Chem. Res., 2015, 54, 2274–2278.
114. Y.-J. Shi, H. Shu, Y.-H. Zhang, H.-M. Fan, Y.-P. Zhang and L.-J. Yang, Fuel Process. Technol., 2016, 150, 141–147.
115. X. Lu, C. Song, C.-C. Chang, Y. Teng, Z. Tong and X. Tang, Ind. Eng. Chem. Res., 2014, 53, 11601–11610.
116. D. Zhang, L. Zhang, L. Shi, C. Fang, H. Li, R. Gao, L. Huang and J. Zhang, Nanoscale, 2013, 5, 1127–1136.
117. X. Lu, C. Song, S. Jia, Z. Tong, X. Tang and Y. Teng, Chem. Eng. J., 2015, 260, 776–784.
118. B. Jiang, Y. Liu and Z. Wu, J. Hazard. Mater., 2009, 162, 1249–1254.
119. Z. Zhu, Z. Liu, S. Liu and H. Niu, Appl. Catal., B, 2001, 30, 267–276.
120. W. Q. Xu, H. He and Y. B. Yu, J. Phys. Chem. C, 2009, 113, 4426–4432.
121. R. Q. Long and R. T. Yang, J. Catal., 1999, 186, 254–268.
122. H. H. Phil, M. P. Reddy, P. A. Kumar, L. K. Ju and J. S. Hyo, Appl. Catal., B, 2008, 78, 301–308.
123. B. Q. Jiang, Z. B. Wu, Y. Liu, S. C. Lee and W. K. Ho, J. Phys. Chem. C, 2010, 114, 4961–4965.
124. J. Zuo, Z. Chen, F. Wang, Y. Yu, L. Wang and X. Li, Ind. Eng. Chem. Res., 2014, 53, 2647–2655.
125. W. Sjoerd Kijlstra, M. Biervliet, E. K. Poels and A. Bliek, Appl. Catal., B, 1998, 16, 327–337.
126. Z. Wu, R. Jin, H. Wang and Y. Liu, Catal. Commun., 2009, 10, 935–939.
127. Z. Liu, Y. Yi, S. Zhang, T. Zhu, J. Zhu and J. Wang, Catal. Today, 2013, 216, 76–81.
128. L. Kylhammar, P.-A. Carlsson, H. H. Ingelsten, H. Grönbeck and M. Skoglundh, Appl. Catal., B, 2008, 84, 268–276.
129. Y. Shi, S. Tan, X. Wang, M. Li, S. Li and W. Li, Catal. Commun., 2016, 86, 67–71.
130. H. Chang, X. Chen, J. Li, L. Ma, C. Wang, C. Liu, J. W. Schwank and J. Hao, Environ. Sci. Technol., 2013, 47, 5294–5301.
131. H. Chang, J. Li, X. Chen, L. Ma, S. Yang, J. W. Schwank and J. Hao, Catal. Commun., 2012, 27, 54–57.
132. Y. Shi, S. Chen, H. Sun, Y. Shu and X. Quan, Catal. Commun., 2013, 42, 10–13.
133. M. D. Amiridis, I. E. Wachs, G. Deo, J.-M. Jehng and D. S. Kim, J. Catal., 1996, 161, 247–253.
134. L. Chen, Z. Si, X. Wu, D. Weng and Z. Wu, J. Environ. Sci., 2015, 31, 240–247.
135. Z. Lei, B. Han, K. Yang and B. Chen, Chem. Eng. J., 2013, 215–216, 651–657.
136. S. Pan, H. Luo, L. Li, Z. Wei and B. Huang, J. Mol. Catal. A: Chem., 2013, 377, 154–161.
137. Z. Sheng, Y. Hu, J. Xue, X. Wang and W. Liao, J. Rare Earths, 2012, 30, 676–682.
138. R. Jin, Y. Liu, Z. Wu, H. Wang and T. Gu, Catal. Today, 2010, 153, 84–89.
139. B. Shen, Y. Wang, F. Wang and T. Liu, Chem. Eng. J., 2014, 236, 171–180.
140. Z. Wu, R. Jin, Y. Liu and H. Wang, Catal. Commun., 2008, 9, 2217–2220.
141. Z. Ma, X. Wu, Y. Feng, Z. Si and D. Weng, Catal. Commun., 2015, 69, 188–192.
142. Z. Sheng, Y. Hu, J. Xue, X. Wang and W. Liao, Environ. Technol., 2012, 33, 2421–2428.
143. B. Shen, T. Liu, N. Zhao, X. Yang and L. Deng, J. Environ. Sci., 2010, 22, 1447–1454.
144. B. Shen, X. Zhang, H. Ma, Y. Yao and T. Liu, J. Environ. Sci., 2013, 25, 791–800.
145. H. Xu, Q. Zhang, C. Qiu, T. Lin, M. Gong and Y. Chen, Chem. Eng. Sci., 2012, 76, 120–128.
146. M. Casapu, O. Kröcher, M. Mehring, M. Nachtegaal, C. Borca, M. Harfouche and D. Grolimund, J. Phys. Chem. C, 2010, 114, 9791–9801.
147. M. Casapu, O. Kröcher and M. Elsener, Appl. Catal., B, 2009, 88, 413–419.
148. B. Yang, D.-H. Zheng, Y.-S. Shen, Y.-S. Qiu, B. Li, Y.-W. Zeng, S.-B. Shen and S.-M. Zhu, J. Ind. Eng. Chem., 2015, 24, 148–152.
149. B. Yang, Y. Shen, S. Shen and S. Zhu, J. Rare Earths, 2013, 31, 130–136.
150. Y. Yu, C. He, J. Chen, L. Yin, T. Qiu and X. Meng, Catal. Commun., 2013, 39, 78–81.
151. M. Pourkhalil, A. Z. Moghaddam, A. Rashidi, J. Towfighi and Y. Mortazavi, Appl. Surf. Sci., 2013, 279, 250–259.
152. J. Huang, Z. Tong, Y. Huang and J. Zhang, Appl. Catal., B, 2008, 78, 309–314.
153. B. Guan, H. Lin, L. Zhu and Z. Huang, J. Phys. Chem. C, 2011, 115, 12850–12863.
154. R. Khodayari and C. U. I. Odenbrand, Appl. Catal., B, 2001, 30, 87–99.
155. L. Chen, J. Li and M. Ge, Chem. Eng. J., 2011, 170, 531–537.
156. R.-T. Guo, Q.-S. Wang, W.-G. Pan, Q.-L. Chen, H.-L. Ding, X.-F. Yin, N.-Z. Yang, C.-Z. Lu, S.-X. Wang and Y.-C. Yuan, J. Mol. Catal. A: Chem., 2015, 407, 1–7.
157. L. Zhang, S. Cui, H. Guo, X. Ma and X. Luo, J. Mol. Catal. A: Chem., 2014, 390, 14–21.
158. F. Moradi, J. Brandin, M. Sohrabi, M. Faghihi and M. Sanati, Appl. Catal., B, 2003, 46, 65–76.
159. A. Zhou, D. Yu, L. Yang and Z. Sheng, Appl. Surf. Sci., 2016, 378, 167–173.
160. R.-T. Guo, Q.-S. Wang, W.-G. Pan, W.-L. Zhen, Q.-L. Chen, H.-L. Ding, N.-Z. Yang and C.-Z. Lu, Appl. Surf. Sci., 2014, 317, 111–116.
161. X. S. Du, X. Gao, R. Y. Qu, P. D. Ji, Z. Y. Luo and K. F. Cen, Chemcatchem, 2012, 4, 2075–2081.
162. S. Boxiong, Y. Yan, C. Jianhong and Z. Xiaopeng, Microporous Mesoporous Mater., 2013, 180, 262–269.
163. B. Shen, L. Deng and J. Chen, Front. Environ. Sci. Eng., 2013, 7, 512–517.
164. A. L. Kustov, M. Y. Kustova, R. Fehrmann and P. Simonsen, Appl. Catal., B, 2005, 58, 97–104.
165. R. Stolle, H. Koeser and H. Gutberlet, Appl. Catal., B, 2014, 144, 486–497.
166. Y. Jiang, X. Gao, Y. Zhang, W. Wu, H. Song, Z. Luo and K. Cen, J. Hazard. Mater., 2014, 274, 270–278.
167. M. Kong, Q. Liu, L. Jiang, F. Guo, S. Ren, L. Yao and J. Yang, Catal. Commun., 2016, 85, 34–38.
168. R.-T. Guo, C.-Z. Lu, W.-G. Pan, W.-L. Zhen, Q.-S. Wang, Q.-L. Chen, H.-L. Ding and N.-Z. Yang, Catal. Commun., 2015, 59, 136–139.
169. W. Li, R.-T. Guo, S.-X. Wang, W.-G. Pan, Q.-L. Chen, M.-Y. Li, P. Sun and S.-M. Liu, Fuel Process. Technol., 2016, 154, 235–242.
170. L. Zhou, C. Li, L. Zhao, G. Zeng, L. Gao, Y. Wang and M. E. Yu, Appl. Surf. Sci., 2016, 389, 532–539.
171. C. Zhu, H. Tian, K. Cheng, K. Liu, K. Wang, S. Hua, J. Gao and J. Zhou, J. Cleaner Prod., 2016, 114, 343–351.
172. P. Wang, S. Su, J. Xiang, H. You, F. Cao, L. Sun, S. Hu and Y. Zhang, Chemosphere, 2014, 101, 49–54.

---