Recent Progress on Improving Low-Temperature Activity of Vanadia-Based Catalysts for the Selective Catalytic Reduction of NO_x with Ammonia

Abstract

Selective catalytic reduction of NO_x with NH₃ (NH₃-SCR) has been successfully applied to abate NO_x from diesel engines and coal-fired industries on a large scale. Although V₂O₅-WO₃(MoO₃)/TiO₂ catalysts have been utilized in commercial applications, novel vanadia-based catalysts have been recently developed to meet the increasing requirements for low-temperature catalytic activity. In this article, recent progress on the improvement of the low-temperature activity of vanadia-based catalysts is reviewed, including modification with metal oxides and nonmetal elements and the use of novel supports, different synthesis methods, metal vanadates and specific structures. Investigation of the NH₃-SCR reaction mechanism, especially at low temperatures, is also emphasized. Finally, for low-temperature NH₃-SCR, some suggestions are given regarding the opportunities and challenges of vanadia-based catalysts in future research.

1. Introduction

Nitrogen oxides (NO_x, including NO and NO₂), primarily emitted from fossil fuel combustion in both stationary and mobile sources, are major pollutants in the atmosphere. NO_x is harmful to human health and can lead to the greenhouse effect, ozone depletion, acid rain, photochemical smog, and haze. Therefore, reducing the emission of NO_x has become one of the most urgent atmospheric environment problems [1,2,3]. Increasingly stringent policies and legislation to control the emission of nitrogen oxides have been enacted all over the world.

In the 1970s, selective catalytic reduction of NO_x with NH₃ (NH₃-SCR) was first applied to NO_x removal from stationary sources in Japan [4], and has subsequently been widely used all over the world for NO_x control from both stationary and mobile sources [5]. The NH₃-SCR process mainly comprises the following reactions:

$$4\mathrm{NO}\left(\mathrm{g}\right)+4{\mathrm{NH}}_3\left(\mathrm{g}\right)+{\mathrm{O}}_2\left(\mathrm{g}\right)\to 4{\mathrm{N}}_2\left(\mathrm{g}\right)+6{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

$$2{\mathrm{NO}}_2\left(\mathrm{g}\right)+4{\mathrm{NH}}_3\left(\mathrm{g}\right)+{\mathrm{O}}_2\left(\mathrm{g}\right)\to 3{\mathrm{N}}_2\left(\mathrm{g}\right)+6{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

$$2\mathrm{NO}\left(\mathrm{g}\right)+4{\mathrm{NH}}_3\left(\mathrm{g}\right)+2{\mathrm{NO}}_2\left(\mathrm{g}\right)\to 4{\mathrm{N}}_2\left(\mathrm{g}\right)+6{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

$$6\mathrm{NO}\left(\mathrm{g}\right)+4{\mathrm{NH}}_3\left(\mathrm{g}\right)\to 5{\mathrm{N}}_2\left(\mathrm{g}\right)+6{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

$$6{\mathrm{NO}}_2\left(\mathrm{g}\right)+8{\mathrm{NH}}_3\left(\mathrm{g}\right)\to 7{\mathrm{N}}_2\left(\mathrm{g}\right)+12{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

Catalysts are the key components in NH₃-SCR technology for abating NO_x emission. Currently, V₂O₅-WO₃/TiO₂ is commercially applied because of its excellent catalytic performance at 300–400 °C and strong SO₂ resistance [6]. Due to the high operating temperature, the precipitator and desulfurizer units should be installed downstream of the SCR unit in power plants [7]. However, this results in deactivation of the catalysts by poisoning with heavy metals, phosphorus and alkali/alkaline earth metals existing in the stack gases. Furthermore, the flue gas temperature from steel, glass and cement plants is lower than the working temperature of traditional V-based catalysts [8]. In order to avoid rapid deactivation and the need for additional energy consumption and to satisfy increasingly stringent NO_x emission limits, the study of novel low-temperature vanadia-based SCR catalysts with high performance is highly desirable and has attracted broad attention to the deNO_x process. Figure 1 shows the numbers of papers published each year from 1991 to 2020 containing “vanadium” and “SCR” in their contents using the “web of science” search engine. The increasing trend in the number of publications reflects the popularity, importance and enthusiasm for research of the vanadia-based catalysts for NH₃-SCR.

Herein, this work reviews the latest progress, especially the progress in the last five years, on vanadia-based catalysts for low-temperature NH₃-SCR, including a summary of methods for improving the performance at low temperature over vanadia-based catalysts and a discussion of the reaction mechanism of low-temperature NH₃-SCR, and finally proposes suggestions for the development of vanadia-based catalysts with superior low-temperature SCR activity.

2. Performance Improvement at Low Temperatures

The narrow operating temperature window of vanadia-based catalysts restricts their broader applications in the deNO_x process. Though catalysts with high loadings of V₂O₅ exhibit high catalytic activity, high vanadia content leads to a decrease in N₂ selectivity and thermal stability and an increase in the catalytic oxidation of sulfur dioxide to sulfur trioxide, which induces severe corrosion problems for equipment [9]. Therefore, researchers have devoted much effort to improving the low-temperature catalytic activity of vanadia-based catalysts, including modification of the active components, supports, and preparation methods, and experimenting with metal vanadates and specific structures.

2.1. The Modification of Vanadia-Based Catalysts

There are many reports on the modification of vanadia-based catalysts with transition metal oxides or rare-earth oxides to promote low-temperature NH₃-SCR activity. V-based catalysts were modified with different transition metal (Co, Mn, Fe, Cu) oxides via the impregnation method, and Cu-V/TiO₂ was found to present the best catalytic performance at 225–375 °C, probably attributable to the increased abundance of active surface oxygen species and strong acid sites [10]. Cu-doped V₂O₅/WO₃-TiO₂ exhibited higher NO_x conversion than a V/WTi sample, which was mainly attributed to the existence of double redox couples of Cu²⁺/Cu⁺ and V⁵⁺/V⁴⁺ from the adjacent copper oxides and vanadium oxides on the surface [11]. In a study of the effects of different CuO doping levels, 7 wt.% Cu/VWTi exhibited high SCR activity and Hg⁰ oxidation at 280–360 °C, resulting from high dispersion of active species and enhancement of the redox properties [12]. In addition, a 0.3 wt.% Ce-0.05 wt.% Cu/VWTi catalyst showed stronger K resistance and higher catalytic activity than the commercial VWTi catalyst by reason of its enhanced surface acidity and redox properties [13].

In the presence of 1000 ppm SO₂ at 280 °C, Ce-doped V/WTi showed excellent catalytic activity. An oligomeric V-O-V structure was formed on the catalyst due to the introduction of ceria, facilitating electronic conduction between the vanadia and ceria species, lowering the apparent activation energy and significantly improving the NH₃-SCR reaction performance. The stability of adsorbed NO_x species was improved, and active vanadium sites were protected from poisoning by the addition of ceria [14]. Ma et al. found that 8 wt.% CeO₂-1 wt.%V₂O₅-WO₃/TiO₂ showed higher than 80% catalytic activity at the temperature range of 190–450 °C, comparable to the performance of 3 wt.% V₂O₅-WO₃/TiO₂, showing that the usage of toxic vanadium oxides can be reduced via ceria doping in the development of low-temperature SCR catalysts [15]. Arfaoui et al. adopted a one-step sol−gel method to synthesize new sulfate and ceria co-modified V₂O₅-TiO₂ nanostructured aerogel catalysts, which presented superior catalytic performance [16]. Hu et al. reported that V-Ce(SO₄)₂/Ti catalysts with abundant reactive acid sites exhibited better activity than the commercial V-W/Ti catalyst and showed a strong tolerance to SO₂ and H₂O [17]. The catalysts 3Ce-V-W/TiO₂ [18], V-5W/Ce/Ti [19], V_0.8WTiCe_0.25 [20] and V₁Ce₁₀Ti [21] all exhibited excellent SCR activity.

Chi et al. adopted an ultrasonic-assisted impregnation method to prepare Ce-Cu modified V₂O₅/TiO₂ catalysts, exhibiting high NO conversion (>97%) at 200–400 °C [22]. Xu et al. found that among TiO₂-supported vanadium oxide catalysts modified with different amounts of Ce and Sb, V₅Ce₃₅Sb₂/Ti presented the best SCR performance, over which 90% NO conversion was obtained at 210 °C [23]. The doping of V and Sb into Ce/Ti increased the concentration of Ce⁴⁺, and V/Sb/Ce/Ti exhibited superior catalytic performance and N₂ selectivity (as shown in Figure 2) [24].

The introduction of Mn into VO_x/CeO₂ {110} can speed up the V⁵⁺/V⁴⁺ redox cycle and increase the amount of coordinated NH₃ species and bridging nitrate species during SCR, and therefore improve the deNO_x activity [25]. Among MO_x-V₂O₅-MoO₃-CeO₂/TiO₂ (M = Mn, Cu, Sb, and La) catalysts, Mn₅V₁Mo₃Ce₇/Ti showed the optimal catalytic activity due to its strong reducibility and abundant acid sites with various strengths, and the tolerance of SO₂ and H₂O were enhanced remarkably by sulfate species formed in the presence of sulfur oxide [26]. V/Mo-Ti presented better NH₃-SCR performance and stronger SO₂ resistance than V/W/Ti catalysts, attributed to the greater abundance of acid sites induced by the introduction of molybdenum, and with increasing Mo⁶⁺ ratio, V/Mo-Ti showed increasing catalytic activity [27]. MoO₃-doped V/WTi exhibited better catalytic performance than V/WTi and 5 wt.% Mo-V/WTi presented the best deNO_x activity because the introduction of MoO₃ enhanced the amount of surface oxygen species and redox capability and altered the number, type and reaction activities of surface acid sites [28]. The doping of Cr into V/TiO₂ enhanced the low-temperature reductive capacity, increased the amounts of acid sites of weak and medium strength, and promoted the rate of the SCR reaction [29]. The optimal catalytic performance was obtained over a 10 wt.% Cr_0.2V_0.8/TiO₂ catalyst. The doping of Nb₂O₅ into the V/WTi catalyst improved the low-temperature deNO_x performance, which was ascribed to the improved dispersion of vanadia and promotion of the reactivity of adsorbed ammonia species [30].

Besides modification with metal oxides, doping with nonmetal elements can also enhance the SCR performance of the catalyst. Marberger et al. found that the doping of 2–4 wt.% SiO₂ into VW/Ti can hamper the growth of anatase crystallites at 600 and 650 °C, resulting in preservation of the low-temperature catalytic activity [31]. Zhao et al. found that the addition of N and S to V₂O₅/TiO₂ can significantly influence NH₃-SCR performance, and a catalyst with 3:1:100 (S:N:Ti) molar ratio exhibited superior performance attributed to the large pore volume and surface area, strong reducibility and increased surface acidity, electronic interactions between TiO₂ and the dispersed vanadia species, and inhibition of the anatase-to-rutile phase transformation [32,33]. The SCR activity at low temperatures was promoted evidently by the doping of F into V₂O₅/TiO₂ because modification with F promoted the interaction of vanadium species and TiO₂ by means of oxygen vacancies with electrons [34]. Liang et al. also found that F-V₂O₅-WO₃/TiO₂ could enhance the SCR activity and the resistance to sulfur and water. The surface morphology of the catalyst was eroded and the grain size was reduced by the introduction of an appropriate amount of F [35]. The introduction of a certain amount of phosphorus into V₂O₅-WO₃/TiO₂ increased the intrinsic activity for the enhancement of Lewis and Brønsted acid sites on V₂O₅-WO₃/TiO₂, and induced the formation of more polymeric surface vanadyl species through spatial effects [36]. Nam et al. added a low content of P into V/Sb/Ce/Ti, and Sb-phosphate and Ce-phosphate instead of VOPO₄ were formed, contributing to the enhancement of the redox properties while keeping the acidity almost unchanged, resulting in the promotion of the SCR activity [37]. A heteropoly acid (HPA)-promoted V₂O₅/TiO₂ catalyst presented better catalytic performance and stronger potassium tolerance than WO₃-promoted catalysts [38]. The modification of vanadia-based catalysts and their catalytic activity are shown in

Table 1. Catalytic performances of vanadia-based catalysts synthesized by different methods.

Catalysts	Preparation Methods	Reaction Conditions	NOx (or NO) Conversion (Temperature Range)	GHSV	Source
Cu-V/TiO2	Impregnation	0.05% NO, 0.05% NH3, 5 vol% O2	>90% (225–375 °C)	26,000 h−1	10
CuV/WTi	Impregnation	0.05% NO, 0.05% NH3, 5 vol% O2	>84% (250–400 °C)	10,000 h−1	11
VWCeCuTi	Impregnation	0.05% NO, 0.05% NH3, 5 vol% O2	>80% (250–375 °C)	60,000 h−1	13
V/Ce/WTi	Precipitation	0.05% NO, 0.05% NH3, 5 vol% O2, 5vol% H2O	95% (250–450 °C)	18,000 h−1	14
VCeTiSO42-	Sol−gel	0.1% NO, 0.1% NH3, 8 vol% O2, 3.5vol%H2O	>80% (275–450 °C)	120,000 h−1	16
V-Ce(SO4)2/Ti	Impregnation	0.08% NO, 0.08% NH3, 5 vol% O2, 0.05%SO2, 5vol%H2O	100% (300–450 °C)	150,000 mL/g·h	17
7%Ce-1%CuV/Ti	Impregnation	0.05% NO, 0.05% NH3, 5 vol% O2	>97% (200–400 °C)	45,000 h−1	22
V5Ce35Sb2/TiO2	Impregnation	0.1% NO, 0.1% NH3, 3 vol% O2	>90% (225–400 °C)	45,000 h−1	23
V-Mn/Ce	Hydrothermal	0.05% NO, 0.05% NH3, 5 vol% O2	>80% (200–350 °C)	160,000 h−1	25
V3Mo5/WTi	Impregnation	0.05% NO, 0.05% NH3, 3 vol% O2	>80% (200–300 °C)	60,000 h−1	28
10%Cr0.2-V0.8/TiO2	Impregnation	0.05% NO, 0.05% NH3, 3 vol% O2	>85% (160–280 °C)	60,000 mL/g·h	29
6%Nb-3% V/W-Ti	Impregnation	0.05% NO, 0.05% NH3, 3 vol% O2	>90% (225–400 °C)	60,000 h−1	30
4%Si2%V/10%W/Ti	Co-impregnation	0.05% NO, 0.06% NH3, 10 vol% O2	>80% (300–500 °C)	50,000 h−1	31
S3N1V/Ti100	Sol−gel	0.05% NO, 0.05% NH3, 5 vol% O2	100% (240–450 °C)	27,549 h−1	32
VTiF-(NH4)2TiF6	Sol−gel	0.05% NO, 0.06% NH3, 5 vol% O2	78.5% (240 °C)	38,900 h−1	34
0.2%F-VW/Ti	Impregnation	0.07% NO, 0.07% NH3, 5 vol% O2	>95% (160–360 °C)	30,000 h−1	35
1P-VWTi	Impregnation	0.05% NO, 0.05% NH3, 5 vol% O2	96% (250 °C)	70,000 h−1	36
5%V/15%TPA/Ti	Impregnation	0.1% NO, 0.1% NH3, 4 vol% O2 2.3 vol% H2O	100% (300 °C)	180,000 h−1	38

.

The improvements in low-temperature performance accomplished via modification with metal oxides or nonmetal elements were mainly related to modulation of the redox capability and acidity of the vanadia-based catalysts. Besides doping with active components, the catalyst supports were also modified to further improve the deNO_x activity.

2.2. The Effect of Different Supports

TiO₂ has been applied as a support for commercial vanadia-based catalysts for decades due to its favorable properties, including good surface dispersion of V species on TiO₂ in comparison with other supports (Al₂O₃, SiO₂, et al.) as well as weak and reversible sulfation under SCR reaction conditions [8,39,40]. However, at high reaction temperatures, anatase TiO₂ will transform to rutile TiO₂, which leads to catalyst sintering accompanied by decreases in the surface area and activity [41]. Recently, many attempts have been made to improve vanadia-based catalysts by modifying the TiO₂ support or using diverse support materials.

Vanadium oxides dispersed on microporous TiO₂ supports produced much less N₂O than commercial TiO₂-supported catalysts during the process of NH₃-SCR reaction, because the formation of bulk-like V₂O₅ species, which resulted in the formation of N₂O, was suppressed (as shown in Figure 3) [42]. The pore structure of TiO₂ determined the types of vanadium species present, which affected the sulfur resistance during the SCR reaction. A 5 wt.%V/Ti (microporous TiO₂) showed stronger resistance to sulfur poisoning and better activity than a 5 wt.%V/Ti (mesopore DT-51) having bulk-like VO_x species [43]. A catalyst with 5 wt.% vanadia ultrasonically impregnated on a TiO₂ support with a large surface area (380.5 m²/g) had a 100 °C wider operating temperature window and higher N₂ selectivity than the traditional vanadia-based catalyst due to enhanced redox capability and total acidity [44]. Ti-bearing blast furnace slag was used to prepare the TiO₂-SiO₂ support, which was then used to support V₂O₅-WO₃ and the catalyst presented enhanced acidity and redox ability for active species and promoted the SCR activity compared with a catalyst using commercial TiO₂ as support [45]. Mixed tungsten−titanium-pillared clays were used to support vanadium oxides, and the synthesized catalysts showed a uniform distribution of WTi-pillars between the clay layers and showed good catalytic performance in NH₃-SCR [46]. Catalysts on other novel supports were also investigated, such as V₂O₅ supported on reduced TiO₂ [47], vanadium impregnated on silica-pillared layered titanate (SiO₂-Ti₄O₉) [48], supported vanadium-substituted Keggin polyoxometalates (POM) [49], V₂O₅/H₂Ti₃O₇-nanotubes and V₂O₅-WO₃/H₂Ti₃O₇-nanotubes [50], 1V4Ce/Ti-PILC (1 wt.% V and 4 wt.% Ce) [51], and V/Ce_1−xTi_x (x = 0.3, 0.5) [52].

Ce-based supports for vanadium oxide NH₃-SCR catalysts have also attracted interest. In our previous study, the homogeneous precipitation method was used to prepare vanadium oxides/CeO₂ catalysts, and the catalyst exhibited high NO_x conversion and strong tolerance to SO₂ and H₂O [53]. Incorporating Ti into V₂O₅/CeO₂ improved the NO_x conversion, N₂ selectivity and resistance to SO₂ and H₂O due to the lower crystallinity, more abundant acid sites, better dispersion of surface V species and greater number of reduced species [54]. Modification of VO_x/CeO₂ with NbO_x promoted the redox capability and acidity, leading to better NH₃-SCR activity and stronger tolerance to SO₂/H₂O than the unmodified VO_x/CeO₂ catalyst; and 30Nb-1VO_x/CeO₂ exhibited better NH₃-SCR performance than 3V₂O_5–10WO₃/TiO₂ (as shown in Figure 4) [55,56].

ZrO₂ supports with various morphologies (mesoporous, rod, star, sphere, hollow) were loaded with vanadium oxide, and 3 wt.% V/Zr (mesoporous) exhibited a wide operating temperature window and excellent N₂ selectivity. The content of tetravalent vanadium ions on the surface of the catalyst and catalytic performance decreased in the order of mesoporous > hollow spheres > stars > rods [57]. In addition, there are several reports on the use of Zr-modified supports such as in 5wt.%V/Zr_0.3Ce_0.7 [58], V/ZrCe_0.6 [59], Co-V/Zr-Ce [60], which enhanced the SCR performance of the vanadia-based catalysts. A catalyst with vanadium−tungsten oxides supported on CuCeO_y microflowers showed excellent deNO_x performance at low temperatures, attributed to the facile electron transfer among V, Cu and Ce ions, decreasing the apparent activation energy of the NH₃-SCR reaction (E_a = 16.59 kJ/mol) [61]. A V/Ce_0.9Fe_0.1 solid solution also presented relatively high SCR activity at low temperature (below 300 °C) [62].

Carbon materials, including carbon nanotubes (CNT), carbon fibers (ACF) and activated carbon (AC), can also be used to support vanadia-based catalysts for NH₃-SCR for large specific surface area and pore volume. Carbon nanotube-supported vanadium oxide (V₂O₅/CNT) exhibited high NO reduction activity and high stability at low-temperatures, and the activity in the presence of sulfur dioxides was significantly promoted [63]. Sulfur dioxide could enhance the catalytic activity of a V₂O₅/SiC catalyst at 250 °C because pre-adsorption with SO₂ + O₂ enhanced NH₃ adsorption and at low temperatures, NO could effectively react with the ammonium bisulfate that was formed on the surface of the catalyst [64]. Excellent SCR performance and SO₂ resistance were also achieved by 3 wt.%Fe-0.5 wt.%V/AC [65], and Mn-Ce-V/AC [66].

Improving the properties of the support could thus improve the dispersion of vanadium species and the interaction between the support and active components, enhance electron transfer between active components, enlarge the specific surface area, and strengthen the redox capability and acidity, all of which are beneficial to accelerating the SCR reaction.

2.3. The Effect of Preparation Method

The preparation methods utilized for catalysts mainly include the precipitation method, sol−gel method, hydrothermal method, impregnation method, and other specialized synthesis methods. Different synthesis methods and synthesis parameters influence the nanostructures, morphologies, and surface physicochemical properties, finally affecting the NH₃-SCR catalytic activity. Hence, it is important to optimize the synthesis method to improve the NH₃-SCR activity over vanadia-based catalysts.

Compared to materials prepared by incipient wetness impregnation, coprecipitated V₂O₅-WO₃/TiO₂ catalysts presented superior SCR performance derived from their increased ammonia adsorption capacity due to the existence of new surface WO_x site-associated surface defects on the TiO₂ support [67]. A V/Ce/WTi-DP (deposition precipitation) catalyst exhibited more O_α, stronger reducibility and more surface Ce species, thus showing less N₂O formation and better medium-temperature NH₃-SCR performance than similar catalysts prepared by impregnation [68]. CeO₂-modified V₂O₅/TiO₂ synthesized via chemical vapor condensation exhibited a higher ratio of Ce³⁺ and showed higher reducibility and acidity than that synthesized by the impregnation method [69]. In our previous work, VO_x/CeO₂ synthesized via the homogeneous precipitation method presented better catalytic performance and stronger tolerance to H₂O and SO₂ than that synthesized via sol−gel method, incipient wetness impregnation and rotary evaporation impregnation, which can be ascribed to the lower crystallinity of CeO₂ on the surface, greater abundance of acid sites and vanadium species, and better dispersion of V species [53]. The support synthesis methods had an effect on structure and performance of VO_x/CeO₂ catalysts, and the catalyst from precipitation method prepared-support was more active than that from citrate method due to higher surface area and a more effective incorporation of V sites into the support surface [70]. A series of VWTi nanoparticle catalysts were directly synthesized via the sol−gel method, and V_0.02W_0.04Ti showed the best SCR activity and the lowest apparent activation energy due to a high concentration of distorted and reducible vanadium species [71].

Besides these conventional synthesis methods, several novel preparation methods were developed to enhance catalytic activity of vanadia-based NH₃-SCR catalysts. Chen et al. prepared V_0.1Ti_0.9O_2-δ catalysts by flame synthesis with promoted low-temperature activity due to strong redox properties, and different synthesis conditions also effected the properties of the catalysts and their catalytic activity [72]. Arfaoui prepared a V₂O₅-CeO₂-TiO₂-SO₄²⁻ nanostructured aerogel catalyst via a one-step sol−gel method accompanied by the supercritical drying process, which presented a large surface area (66 m²/g), large porosity and good thermal stability [16]. Tuning the interaction degree supplied by external forces can regulate the morphology. Rotation-assisted hydrothermal synthesis can be used to prepare thick multiwalled titanate nanotubes and when used as the support for vanadium−tungsten-oxide, the resulting catalysts presented stable catalytic activity because the sintering of VO_x was suppressed by the multiwalled nanotubes [73]. Doronkin et al. found that V/Ti and V-W/Ti catalysts prepared via incipient wetness impregnation and grafting with highly dispersed, isolated and polymeric V-oxo species exhibited high SCR activity [74].

In the synthesis of catalysts, different preparation parameters can also affect the physicochemical properties and the catalytic activity. For example, different vanadium precursors have an influence on the dispersion of V species and catalytic performance [75]. V₂O₅/TNTs catalysts prepared using VOSO₄ as the V precursor exhibited stronger synergistic effects with the TNTs than those made with a NH₄VO₃ precursor, and active metal precursors containing cation groups were superior to those with anion groups (as shown in Figure 5) [76]. During the preparation process of iron vanadate, the pH value determined the stoichiometry of the FeVO₄/TiO₂-WO₃-SiO₂ product: vanadium-rich samples were prepared at pH lower than 6, stoichiometric FeVO₄ was obtained in the pH range of 4–6, while samples prepared at pH > 6 were rich in Fe₂O₃ [77]. The NH₃ adsorption capacity of Ce-doped V₂O₅-WO₃/TiO₂ was affected by the sequence of impregnation by Ce and W, and 0.2V-5W-5Ce/Ti exhibited stronger SO₂ and H₂O tolerance than 0.2V-(5Ce5W)/Ti and 0.2V-5Ce-5W/Ti [78]. Compared to the untreated sample, V-Ce-Ni/TiO₂ catalysts treated by nonthermal plasma possessed smaller particle size, better dispersion and more uniform distribution of active sites, as well as larger number of oxygen vacancy and acid sites. Therefore, the plasma-treated catalysts showed improved catalytic performance at low temperatures with a wider operating temperature window [79]. Moreover, the treatment atmosphere can also have an influence on catalytic performance. The ratios of active polymeric V-O-V species and further NH₃-SCR activity decreased in the following order: V/Ti > V/Ti-O₂ > V/Ti-N₂ > V/Ti-NH₃ [80].

2.4. Metal Vanadates

Although vanadia-based catalysts show superior catalytic performance at medium temperatures and have been widely applied in the deNO_x process, the volatility and the toxicity of vanadium oxide at high temperature is a crucial problem. However, metal vanadates present better thermal stability than V₂O₅ due to their higher melting points (e.g., 780, 850, 1030 and 1100 °C for copper vanadate, iron vanadate, manganese vanadate and cerium vanadate, respectively). Therefore, metal vanadates can be candidates for vanadia-based catalysts with great potential [81].

Iron vanadates (FeVO₄) have recently been extensively studied as active NH₃-SCR catalysts. Abundant surface defects existed on a FeVO₄/TiO₂ catalyst for the adsorption and activation of reactants, and the true active sites were FeVO₄ phase surface-enriched with VO_x species. Thus, similar to V₂O₅/TiO₂, FeVO₄/TiO₂ showed superior catalytic performance and H₂O/SO₂ tolerance [1]. Fe_0.75V_0.25O_δ exhibited excellent SCR performance in the temperature range of 175–400 °C. It was found that the formation of amorphous FeVO₄ resulted from the incorporation of V into Fe₂O₃, and the apparent activation energy decreased due to the synergistic effect of FeVO₄ and Fe₂O₃ improving the catalytic activity at low temperatures [82]. Due to the greater electronic inductive effect, Fe₂V₄O₁₃/TiO₂ showed stronger redox capability and more sites accessible to NO_x/NH₃ than FeVO₄/TiO₂ and thus presented higher activity in the presence of H₂O [83].

Marberger et al. found that compared to the 2.3 wt.% V₂O₅/TiO₂-WO₃-SiO₂ catalyst, 4.5 wt.% FeVO₄/TiO₂-WO₃-SiO₂ showed enhanced catalytic performance, and the decomposition of FeVO₄ led to an activation effect due to the dispersal and migration of VO_x species to the surface of the support material, which were the active species responsible for NH₃-SCR [84]. Wu et al. proposed that the performance of FeVO₄/TiO₂-WO₃-SiO₂ was structure-sensitive, and that the pH values adopted during the preparation process had an influence on the nanostructure and morphology. The optimal catalyst showed 90% NO_x conversion in the temperature range of 246–476 °C and strong tolerance to SO₂ and H₂O [85]. The doping of Er into FeVO₄/TiWSi could block the transformation to rutile, inhibiting the deactivation of FeVO₄ due to thermal aging and thus improving the activity after thermal treatment, and the Fe loading determined the medium/low-temperature catalytic activity. Fe_0.5Er_0.5VO₄ was found to exhibit superior catalytic performance and stability [86].

The oxidation of NO to NO₂ was enhanced by the coexistence of Ce⁴⁺ species stabilized as CeO₂ with bulk CeVO₄. Aging treatments under wet atmosphere at 500 and 600 °C did not result in the sublimation and loss of vanadium [87]. The introduction of Zr into CeVO₄ to form a Ce_1−xZr_xVO₄ solid solution led to high activity, with a 125 °C light-off temperature and a wide temperature window of 150–375 °C by reason of the increased surface area, surface active oxygen species and acid sites of the catalysts [81]. Zr-CeVO₄/TiO₂-nanosheets showed better SCR performance, stability and tolerance to H₂O/SO₂ than nanoparticles due to their more abundant Brønsted acid sites and active oxygen species [88]. A catalyst with Ce:V = 1:1 supported on sulfated zirconia showed high activity in the presence of SO₂ and potassium. The incorporation of vanadium led to the formation of CeVO₄, preventing reaction between SO₂ and CeO₂ and maintaining the reactivity of active sites, thus enhancing the tolerance toward SO₂ [89]. The doping of Sn into CeVO₄ catalysts can broaden the temperature window and improve the resistance to SO₂ and H₂O, mainly resulting from the large specific surface area (from 40.75 to 49.05 m²/g), strong interactions among vanadium, cerium and tin, and large numbers of oxygen vacancies and acid sites [90].

Compared to CuV₂O₆ (Cu₁/Ti), Cu₂V₂O₇ (Cu₂/Ti), and Cu₃V₂O₈ (Cu₃/Ti), Cu₅V₂O₁₀ (Cu₅/Ti) presented optimal redox behavior and the largest number of acid sites, and therefore presented the greatest SCR performance at low temperatures [91]. The addition of Sb as a promoter can further improve the SCR activity over copper vanadate catalysts due to the improved redox properties [91]. SO₂ and O₂ can be used to modify Sb-promoted Cu₃V₂O₈ on TiO₂, and 400 °C was an adequate functionalization temperature to promote the redox behavior and increase the number of Brønsted acid sites, resulting from the more abundant surface Cu(SO₄) or from a proper combination of the metal-bound SO_Y²⁻ species with monodentate and bidentate binding configurations, to enhance the catalytic performance (as shown in Figure 6) [92].

In summary, metal vanadates can not only increase the low/medium-temperature NH₃-SCR activity but also enhance the thermal stability [93,94]. The metal vanadates have aroused the interest of researchers due to their excellent chemical and physical properties. In future studies, the metal vanadate can obtain higher SCR activity through doping different elements or changing the preparation methods, etc.

2.5. Specific Structures

Tuning the structure and morphology of nanoparticle catalysts can enhance the NH₃-SCR activity. Thus, researchers have made great efforts to regulate the structure and morphology of vanadia-based catalysts. A FeVO₄ nanorod/TiO₂ monolith catalyst exhibited a remarkably higher catalytic activity than a traditional FeVO₄ nanoparticle/TiO₂ catalyst, due to the predominantly exposed reactive planes (−210) contributing to the stronger redox capability and more abundant surface active oxygen species [95]. Multichannel TiO₂ nanotubes can provide abundant surface-adsorbed oxygen species and anchor active components efficiently, and a CeVTi-nanotube catalyst presented satisfactory NH₃-SCR activity in the temperature range of 220–460 °C [96]. Bulk TiO₂ was treated by a hydrothermal reaction to obtain zeolitic microporous TiO₂ to support vanadia-based catalysts, and compared to conventional V₂O₅/TiO₂, the catalyst not only maintained excellent SCR performance but also suppressed N₂O emission significantly [97]. V/CeO₂ nanopolyhedrons showed better SCR activity than nanocubes and nanorods for their abundant surface acid sites and appropriate redox ability (as shown in Figure 7) [98]. TiO₂ with different crystal types was used to support 3 wt.%V₂O₅, and the catalytic activity was found to depend on the dominant crystal facets of the TiO₂ nanoparticles. V₂O₅ loaded onto sheet-like TiO₂, on which anatase (001) facets were preferentially exposed, presented a better catalytic performance than that loaded onto commercial TiO₂ (TiO₂-P25) or octahedral TiO₂ with preferentially exposed anatase (101) facets, as the catalyst presented larger amounts of chemisorbed oxygen and better dispersion and stronger reducibility of V species [99]. A titania nanotube-encapsulated vanadium oxide catalyst exhibited a monolayer of isolated species where V⁵⁺ was the dominant oxidation state and demonstrated a wide operating temperature window, which was due to the active species being well dispersed on the support surface [100].

3. Reaction Mechanism at Low Temperatures

Although vanadia-based NH₃-SCR catalysts have been commercialized for decades, the reaction mechanism on the catalysts has still been a focus of research in recent years, including the identification of active sites and intermediates and elucidation of the interaction of the active sites with the reactants. The NH₃-SCR reaction pathway proceeds as follows: the adsorbed NH₃ species react either with adsorbed nitrites/nitrates (Langmuir−Hinshelwood mechanism, i.e., L–H) or directly with gaseous NO (Eley−Rideal mechanism, i.e., E–R) to generate intermediates that subsequently decompose to N₂ and H₂O [5,101,102]. The acid sites and redox sites work together in the NH₃-SCR reaction on the catalysts, and the close coupling of acid and redox sites is a design principle for SCR catalysts with excellent NO_x purification efficiency [103].

In a debating issue about the active sites of NH₃-SCR on VWTi catalysts, it has been reported before that both surface NH₄⁺_,ads and NH_3,ads participated in the NH₃-SCR reaction and the coverage of surface tungsten and vanadia oxide species, moisture, and temperature determined their relative population [104]. The large amount of surface NH₄⁺_,ads intermediates present a lower specific SCR activity (TOF) than the minority surface NH_3,ads intermediates. The SCR reaction rate does not depend on the exposed Ti⁴⁺ sites on the support. Marberger et al. [105] proposed that NO reacts predominantly with coordinated NH₃ adsorbed on Lewis acid sites consisting of isolated V⁵⁺ which were reduced only in the coexistence of NH₃ and NO, and the reduction of V⁵⁺ was accompanied by the formation of a nitrosamide intermediate at low temperature. Brønsted acid sites, serving as an NH₃ pool and hardly contributing to the NH₃-SCR reaction, can replenish the Lewis acid sites. The rate-determining step is re-oxidation and regeneration of active Lewis acid sites.

The presence of H₂O and changes in the redox state of vanadia-based catalysts led to a reversible change between Brønsted and Lewis acid sites. Under reducing conditions, the surface was enriched with Lewis acid sites because that ammonia could be coordinated on the reduced vanadium sites as supplementary of Lewis acid sites, while more abundant Brønsted acid sites on the V/Ti catalyst surface were generated under an oxidizing environment [106]. Brønsted acid sites are important at low temperatures, while Lewis acid sites dominate the overall reaction at high temperatures. Lewis acid sites can transform to Brønsted acid sites at temperatures above 300 °C through hydrogen migration [107]. The number of Lewis acid sites decreased and the amount of Brønsted acid sites increased with changes in the vanadium content. The increase of vanadium pentoxide loading resulted in the increase of the amount of V-OH in polymeric vanadyl and the decrease of the amount of isolated vanadyl (V=O) species [108].

Most current investigations of the NH₃-SCR reaction process over V-based catalysts have mainly concentrated on isolated monomeric vanadyl species, while the polymerization of vanadium species and the coupling effects were not taken into consideration. In our previous study [109], DFT calculations were carried out to elucidate the differences between the NH₃-SCR mechanisms taking place on polymeric and monomeric vanadyl species at atomic scale. The results showed that NH₃ adsorbed on surface Ti sites transfered a hydrogen atom to the vanadyl species firstly and then reacted with gaseous NO according to the Eley−Rideal mechanism. An NH₂NO intermediate and a V-OH or V-OH₂ group were formed and then N₂ and H₂O were generated from the decomposition of NH₂NO. The consumed surface oxygen on the vanadyl species was replenished by gas-phase O₂. The catalytic cycle was completed when V=O groups were regenerated (as shown in Figure 8).

For the monomeric vanadyl species, a VOOH intermediate was formed after the transfer of an H atom of the adsorbed NH₃ to the adsorbed O₂ on vanadyl species and rapidly transformed into O=V-OH. However, for polymeric vanadyl species, the thermal stability and lifetime of the VOOH intermediate were enhanced because a hydrogen bond was formed between VOOH and the adjacent V=O group. Therefore, on polymeric vanadyl species, the VOOH intermediate can easily react with NO. The overall reaction barrier of the catalytic cycle decreases and the reaction pathway for the regeneration of redox sites is shortened due to the coupling effect of polymeric vanadyl species. Therefore, the NH₃-SCR reaction rate is significantly enhanced on polymeric species.

Jaegers et al. clarified that the formation of oligomeric vanadia structures on supported V₂O₅-WO₃/TiO₂ was promoted by the unreactive surface tungsten oxide (WO₃), revealing a 2-site mechanism due to the presence of a proportional relationship of SCR reaction rate to the square of surface VO_x concentration. The enhancement of the NH₃-SCR reaction from the incorporation of a promoter occurs via a structural effect generated by adjacent surface sites rather than an electronic effect [110].

The presence of H₂O can have an influence on the reaction pathway for NH₃-SCR at low temperatures. In the standard and fast SCR reactions, adsorbed NH₃ reacts with nitrite species to generate N₂ and H₂O (nitrite path), while the reaction between gaseous NO and NH₄NO₃ formed from adsorbed ammonia species and surface nitrates to emit NO₂, H₂O and N₂ (NH₄NO₃ path) does not occur at low temperatures without H₂O, and in the presence of H₂O both the “nitrite path” and the “NH₄NO₃ path” contribute to the NH₃-SCR reaction [111].

For vanadia-based catalysts, different NH₃-SCR reaction pathways occur on various vanadium species. The reaction mechanism can be more complicated over modified vanadia-based catalysts and is also related to the concentration of vanadium oxide and the reaction temperature. Therefore, a variety of modern characterization and computational techniques should be applied in order to have a better view of the NH₃-SCR reaction mechanism, guiding the design of excellent deNO_x catalysts.

4. Conclusions and Perspective

Commercial V₂O₅-WO₃/TiO₂ catalysts have been widely applied for abating nitrogen oxides from mobile and stationary sources. However, the poor low-temperature activity of vanadia-based catalysts restricts their broader application in the process of NO_x purification. Therefore, this review summarizes the latest research progress on the improvement of deNO_x performance over vanadia-based catalysts at low temperatures and the NH₃-SCR reaction mechanism, and outlines opportunities and challenges for vanadia-based catalysts in the near future.

The NH₃-SCR catalytic performance over vanadia-based catalysts at low temperatures is mainly determined by polymeric vanadyl species, while monomeric species presented lower catalytic activity. The NH₃-SCR reaction needs acid sites and redox sites to work together. The good dispersion of sites with the same function and close coupling of sites with different functions are crucial for the design of excellent NH₃-SCR catalysts. However, considering the complexity of the SCR reaction, the NH₃-SCR reaction mechanisms and the structure−activity relationships still need to be investigated in-depth to gain an enhanced understanding and thus improve catalyst design theory. Furthermore, due to the multielement composition of the developed catalysts, the synergistic reaction mechanisms between different active components, additive and diverse supports need in-depth exploration.

The modification of vanadia-based catalysts with metal oxides or nonmetal elements can improve the low-temperature SCR activity effectively. Diverse supports have been used to improve the dispersion of vanadium oxide and the interaction between the supports and the active components. The utilization of different preparation methods and use of metal vanadates and specific structures can also promote the catalytic performance of vanadia-based catalysts at low temperatures. Though great progress has been made recently, the catalysts still face many challenges in practical application, such as alkali poisoning and halogen poisoning. We need to develop catalysts with better performance to deal with the practical problems with superior low-temperature activity, wide operation windows, strong mechanical and thermal stability, and excellent resistance to alkali, heavy-metal, SO₂, and P/HCl poisoning. However, due to the volatility and toxicity of vanadium oxide, in the future catalysts with a low loading of vanadium or even no vanadium may be the main development trend in the field of NO_x removal.