Research progress on the deactivation mechanism and deactivation inhibition strategy of Rh-based catalysts in exhaust gas treatment

Abstract

With the acceleration of industrialization and the increase in car ownership, exhaust pollution has become a major global environmental challenge. Harmful gases such as carbon monoxide (CO) and nitrogen oxides (NO_x) are discharged into the air, forming acid rain, producing chemical smog, and destroying the ozone layer. This causes severe harm to the environment and human health. Rh-based catalysts play a crucial role in exhaust gas treatment due to their excellent performance in selective catalytic reduction (SCR) and direct nitrogen oxide decomposition (DND) of NO_x. However, deactivation limits their stability and service life, increasing costs and restricting industrial application. This paper reviews the reaction mechanisms of Rh-based catalysts in SCR and DND reactions, discusses deactivation mechanisms, and proposes improvement strategies. It provides theoretical basis and practical guidance for the development of efficient and stable catalysts for exhaust gas treatment.

---

1. Introduction

With the acceleration of industrialization and global economic development, human society faces increasingly severe environmental pressures. The problem of exhaust gas pollution driven by expanding industrial emissions and surging vehicle ownership has become a major global environmental challenge. Harmful gases such as carbon monoxide (CO), nitrogen oxides (NO_x), and volatile organic compounds (VOCs) in these emissions pose significant threats to environmental integrity and human health. Consequently, developing efficient exhaust treatment technologies is imperative for environmental protection. Catalysts play a pivotal role in this process, making research on efficient and eco-friendly catalysts a critical focus. In the field of exhaust aftertreatment, catalysts are primarily categorized into two types based on the engine type and operational principles: three-way catalysts (TWCs) and diesel oxidation catalysts (DOCs). TWCs are predominantly used in gasoline engines, which operate under near-stoichiometric air–fuel ratio conditions. They simultaneously catalyze the reduction of NO_x, and the oxidation of CO and unburned hydrocarbons (HCs). In contrast, DOCs are employed in diesel engines, which operate under lean-burn (oxygen-rich) conditions. The primary function of DOCs is to oxidize CO and HCs, while the reduction of NO_x typically requires additional systems, such as the selective catalytic reduction (SCR) unit. These two types of catalysts operate under distinctly different reaction atmospheres (stoichiometric or lean burn), which also results in significant differences in their deactivation mechanisms. Current common catalysts include metal oxide,^1,2 perovskite,^3,4 molecular sieve,^5–7 and nitrogen doped carbon materials,^8–11 which are classified according to the support. The most critical substance of the catalyst is the active component, and the commonly used active components are precious metals.^12–14 In general, catalysts with precious metals as active components are also called precious metals. Although noble metal catalysts are costly and susceptible to aggregation and deactivation at high temperatures, their strong poison resistance and high thermal stability render them indispensable for industrial exhaust treatment.¹⁵

According to literature reports, noble metals platinum (Pt), rhodium (Rh) and palladium (Pd) are widely used in noble metal catalysts for degrading harmful exhaust gases.¹⁶ Rh not only catalyzes methane oxidation but also exhibits excellent performance in NO_x decomposition. The partially filled 4d orbital of Rh facilitates the adsorption and dissociation of reactants, promotes moderate adsorption, and contributes to intermediate formation.¹⁷ Furthermore, Rh-based catalysts demonstrate significant potential for treating carbon monoxide (CO) and hydrocarbons (HCs).^18,19 These properties establish Rh as an indispensable component in exhaust gas treatment systems.^20,21 However, in practical applications, catalyst deactivation induced by high temperatures or complex reaction atmospheres severely compromises catalytic efficiency and lifespan. To elucidate the deactivation behavior of catalysts under various operating conditions, researchers conducted systematic studies on the catalyst deactivation process, identified key deactivation factors, and thoroughly investigated and clarified the corresponding deactivation mechanisms. At present, the deactivation mechanisms of Rh catalysts are mainly divided into two types, one is chemical poisoning caused by chemical factors, and the other is physical poisoning caused by physical factors.

Chemical poisoning of Rh-based catalysts refers to the phenomenon that some chemical substances occupy or destroy the active center of the catalyst through strong chemical adsorption or chemical reaction with the active Rh site, leading to significant degradation of catalytic activity, selectivity, or stability. Chemical poisoning includes sulfur poisoning, NO_x poisoning, H₂O poisoning, etc. Sulfur poisoning involves deactivation by sulfur-containing compounds via chemical adsorption, reaction, or physical blockage of active sites. Chang et al.²² demonstrated the effect of SO₂ on the NO removal performance of Rh/Al₂O₃ and Rh–Na/Al₂O₃ catalysts under simulated waste incineration conditions. The experimental results showed that when SO₂ is present in the exhaust gas, the Rh/Al₂O₃ catalyst is obviously deactivated. Fig. 1a shows the CH₄ conversion rate of a Rh/ZMS-5 catalyst after adding different amounts of SO₂ to the reaction atmosphere. After adding SO₂ to the reaction atmosphere, the performance of the catalyst has a downward trend, and the performance of the catalyst after adding 10 ppm SO₂ has the most serious decline. This shows that when there is SO₂ in the reaction atmosphere, the ignition temperature of the catalyst will move to the high temperature range, and T₉₀ will also increase, which will seriously affect the catalytic performance of the catalyst. Fig. 1b shows the comparison of CO conversion rates of four Ce-based catalysts loaded with different noble metals after the introduction of 500 ppm H₂S. Among them, the performance of the RhCe catalyst decreased the most seriously, and the conversion rate decreased from about 100% to about 50% at the same time, indicating that Rh has the weakest sulfur resistance compared with the other three noble metals and is more vulnerable to sulfur poisoning. NO_x poisoning refers to the phenomenon that the activity, selectivity, or stability of catalysts is reduced due to the interaction between catalysts and nitrogen oxides (NO_x including NO, NO₂, etc.) or their derivatives (such as O and N after NO decomposition). NO_x and its reaction products may destroy the active site or structure of the catalyst through chemical adsorption, side reactions that produce inert compounds or physical blockage, thus causing deactivation. The research on NO_x poisoning mainly focuses on the adsorption and reaction of NO_x on the catalyst surface. Unlike sulfur poisoning, NO_x acts as both a reactant and poison, requiring mechanistic studies to distinguish its dual roles. H₂O poisoning arises when H₂O competitively adsorbs on active sites, induces chemical transformations, physically blocks pores, or alters support properties. Fig. 1c compares the catalytic performance of a Rh/CeO₂ catalyst after adding H₂O to the reaction atmosphere. After adding H₂O, the catalyst shows obvious H₂O poisoning. At about 350 °C, the conversion of N₂O decreased by about 60% after H₂O poisoning of the fresh catalyst, while that of the aged catalyst decreased by 40%. Obviously, the impact of H₂O poisoning on the Rh-based catalyst cannot be ignored. Although the deactivation of the Rh-based catalyst due to H₂O poisoning is relatively rare, H₂O is one of the components of the reaction atmosphere in most cases, so the influence of H₂O cannot be ignored.

Fig. 1 a: Sulfur poisoning of 2 wt% Rh/ZSM-5 at different concentrations of SO₂ (ref. 25) (Copyright 2020 ELSEVIER). b: The variation of CO conversion rate with running time for CeO₂ catalysts loaded with different precious metals²⁶ (Copyright 2024 ELSEVIER). c: Deactivation of Rh/CeO₂ by water poisoning²¹ (Copyright 2020, ELSEVIER).

In addition to poisoning, the chemical changes on the catalyst surface are also one of the important reasons for deactivation. The surface chemical change is usually the chemical structure change or composition change on the surface of the catalyst itself, which is caused by the oxidation/reduction environment or reaction by-products. Almusaiteer et al.²³ explored the surface chemical behavior of a Rh catalyst in a NO–CO reaction by in situ diffuse reflectance infrared Fourier transform spectroscopy (in situ DRIFTS) combined with surface state analysis. It was found that a high NO/CO ratio or a high concentration of oxidants would promote the oxidation of Rh⁰ to Rh⁺, resulting in the reduction of NO activity. In addition, it was found that a variety of components (O₂, H₂, C₃H₈, NO, and CO) showed competitive adsorption characteristics in the simulated exhaust gas experiment, which together led to the reduction of catalyst activity. Machida et al.²⁴ studied the thermal behavior of Rh loaded Al₂O₃ under a high temperature reduction atmosphere. It was found that when Rh/Al₂O₃ was aged at 900 °C in an air atmosphere, Rh₂O₃ would react with the Al₂O₃ support to form a thermally stable RhAlO_x mixed oxide phase, resulting in a large reduction of metal reaction sites on the surface and a reduction of NO activity of the catalyst.

Physical deactivation of catalysts involves structural or morphological changes that degrade catalytic activity. The primary mechanisms include: (1) sintering deactivation. At elevated temperatures (typically >600 °C), thermal activation triggers Rh particle agglomeration, reducing the active surface area and altering the catalyst structure.²⁷ Ikeda et al.²⁸ studied the thermal stability and catalytic activity of a SiO₂ coated Rh catalyst for NO/CO reactions. After aging (900 °C, 12 h), BET results showed that the specific surface area of the Rh catalyst coated with SiO₂ decreased from 246 m² g⁻¹ to 170 m² g⁻¹, and the specific surface area of a Rh catalyst prepared by a sol–gel method decreased from 231 m² g⁻¹ to 34 m² g⁻¹. The T_50/NO of the former increased by about 20 °C, while the T_50/NO of the latter increased from 320 °C to 420 °C. High temperature aging further exacerbated pore blockage and structural collapse. The essence is still that the sintering phenomenon at high temperature led to the deactivation of the catalyst. Fig. 2a and b explore the change rule of CO oxidation and NO reduction conversion of a Pt/Rh catalyst under different aging conditions (temperature and time). It can be seen from the figure that the aging temperature and time have a great impact on the performance of the catalyst. The NO conversion rate of the catalyst aged at 1200 °C was close to 0%, and it was basically completely deactivated. (2) Active site encapsulation. In addition to the sintering of the catalyst caused by high temperature (mainly represented by a significant reduction in specific surface area), Rh particles will be partially wrapped by the support to form an “encapsulated” structure under the action of specific external driving forces (such as a strong reducing atmosphere). Bernal et al.²⁹ confirmed that H₂ reduction at 900 °C induces CeO₂ overgrowth on Rh nanoparticles in a Rh/CeO₂ catalyst. Crucially, the metal loading concentration additionally modulates structural stability.

Fig. 2 a: CO conversion rates of Pt/Rh under different aging conditions; b: NO conversion rates of Pt/Rh under different aging conditions³⁰ (Copyright 2010, ELSEVIER).

In summary, this paper reviews the research progress of Rh catalyst deactivation, with subsequent sections focusing on the reaction mechanism, deactivation mechanism and improvement strategy of Rh catalysts.

2. Reaction mechanism

2.1 Reaction type

2.1.1 SCR reaction. Selective catalytic reduction (SCR) is a primary NO_x control technology at present.³¹ It utilizes reductants (such as NH₃, urea, or hydrogen) to selectively convert NO_x into N₂ and H₂O over catalysts. The technology using NH₃ as a reductant is called NH₃-SCR, the technology using H₂ as a reductant is called H₂-SCR, and the technology using CO as a reductant is called CO-SCR technology. Due to the oxygen contained in the exhaust gas of a fixed source and a mobile source, it is difficult to selectively catalytically reduce NO_x, which needs to be solved by designing specific catalysts, which is also key to SCR technology. Understanding these mechanisms is crucial for developing efficient catalysts. Subsequent sections elaborate on these technologies through their characteristic chemical reactions.
(1) NH₃-SCR technology. NH₃-SCR technology is the most widely used and mature technology at present.³² The main reactions involved in NH₃-SCR technology are as follows:³³

4NO + 4NH₃ + O₂ → 4N₂ + 6H₂O (1)

2NO₂ + 4NH₃ + O₂ → 3N₂ + 6H₂O (2)

NO + NO₂ + 2NH₃ → 2N₂ + 3H₂O (3)

To intuitively understand the mechanism of NH₃-SCR in a specific reaction, Fig. 3 shows the cycle mechanism of the NH₃-SCR reaction on a Cu-SSZ-13 catalyst. The reaction takes Cu species (oxidation–reduction between Cu²⁺ and Cu⁺) as the core active center and is divided into two parts: oxidation half cycle and reduction half cycle. In the reduction half cycle, Cu²⁺ combines with NH₃ to form Cu²⁺–NH₃ species, then reacts with NO, and finally generates N₂ and H₂O through the transformation of intermediate species. At the same time, Cu²⁺ is reduced to Cu⁺, and some of the Cu⁺ can be oxidized to Cu²⁺ with the participation of H⁺. In the oxidation half cycle, Cu²⁺ can interact with species such as NO₂⁻/H⁺ and NO₃⁻/H⁺, or Cu²⁺ reacts with NO₂ (generated by the reaction of NO and O₂) to form intermediate species (such as Cu²⁺–NH⁴⁺NO₂⁻). These intermediate species further react to form N₂ and H₂O, while maintaining the valence cycle of Cu species to ensure the continuous reaction. This figure specifically shows the whole process of NH₃-SCR on Cu-SSZ-13. Cu²⁺ acts as the active center and adsorbs and dissociates gas molecules, and the zeolite framework protects the active center from being affected. And it can be seen from the circulation diagram that NH₃, as a reductant, needs to be continuously added in the reaction process to maintain the continuous progress of the reaction, which is also an important factor limiting the reaction.

Fig. 3 NH₃-SCR reaction on a Cu-SSZ-13 catalyst³⁴ (Copyright 2023, ELSEVIER).

Although NH₃-SCR technology occupies a dominant position in the field of diesel vehicle exhaust purification with its mature catalytic system and efficient NO_x removal ability, its dependence on the ammonia source (urea) also brings challenges such as ammonia escape, low temperature activity limitation and urea storage and transportation cost.

(2) H₂-SCR technology. In recent years, with the rapid development of hydrogen energy technology and the promotion of the goal of “carbon neutralization”, a cleaner and more forward-looking alternative, H₂-SCR, has gradually entered the research field of vision. The main reactions involved in H₂-SCR technology are as follows:

2NO + 4H₂ + O₂ → N₂ + 4H₂O (4)

2NO + 3H₂ + O₂ → N₂O + 3H₂O (5)

2NO + H₂ → N₂O + H₂O (6)

2NO + 5H₂ → 2NH₃ + 2H₂O (7)

2NO + O₂ → 2NO₂ (8)

2H₂ + O₂ → 2H₂O (9)

Among these, reaction (4) is the most ideal green reaction, and its products are harmless to the environment. Reactions (5) and (6) produce strong greenhouse gas N₂O, leading to secondary pollution. In reactions (7) and (8), the NH₃ and NO₂ generated play a synergistic catalytic role, respectively, and the activity of NO₂ is higher than that of NO, which is beneficial to the reaction process. Reaction (9) is a combustion reaction of H₂, which competes with reaction (4), resulting in a decrease in efficiency.^35–37

(3) CO-SCR technology. CO-SCR technology was first proposed in 1965. The unique feature of this technology is the combination of CO treatment and NO_x emission reduction: the CO in the exhaust gas that originally needs to be treated is used as a reducing agent to replace the ammonia source in the traditional NH₃-SCR technology. Due to the stable chemical properties and strong reduction ability of CO, this double effect treatment mode of “treating waste with waste” has attracted extensive attention in recent years. The core reaction mechanism is to reduce NO_x and oxidize CO to CO₂ at the same time through the role of the catalyst, so as to realize the efficient collaborative removal of NO_x and CO.³⁸ The main reactions involved in CO-SCR technology are as follows:

NO_(g) → NO_(ads.) (10)

CO_(g) → CO_(ads.) (11)

NO → N_(ads.) + O_(ads.) (12)

2N_(ads.) → N_2(g) (13)

N_(ads.) + NO_(ads.) → N₂O_(ads.) (14)

N₂O_(ads.) → N_2(g) + O_(ads.) (15)

N₂O_(ads.) → N₂O_(g) (16)

O_(ads.) + CO_(ads.) → CO_2(g) (17)

The reaction process involves four parts: the adsorption of reactant molecules, the dissociation of adsorbed molecules, the recombination of surface active substances, and the desorption of product molecules.³⁹ NO and CO are first adsorbed on the catalyst surface and activated. The key step is the dissociation of adsorbed NO to form N and O radicals (the decisive step), in which O is used for oxidizing the active center. Subsequently, N is transformed through two ways: (1) pairwise combination to form N₂; (2) reaction with adsorbed NO to form N₂O (some of which is further dissociated to N₂). At the same time, the adsorbed CO reacts with the oxidized active site to generate CO₂, while reducing the active site to maintain the catalytic cycle.

2.1.2 DND reaction. Compared with conventional SCR, direct NO decomposition (DND) has attracted much attention because it does not need to use toxic reductants or external reductants (such as ammonia), and is the most ideal, economic and environmentally friendly method to realize the catalytic conversion of nitrogen oxides. Considering the environmental benefits and economic costs, the catalyst system for direct decomposition of NO_x shows significant advantages. At the beginning of the 20th century, relevant studies showed that NO could decompose directly on the Pt surface above 800 °C. However, due to the low activity of the catalyst and the high cost of materials, this discovery saw gradually limited practical adoption.⁴⁰ However, studies in the early 1990s found that catalyst systems containing copper ion-exchange zeolites, perovskite oxides and noble metals have shown high activity in the direct decomposition of NO_x, which prompted the academic community to invest a lot of energy in the research of NO_x direct decomposition catalysts. In recent years, the research of NO_x direct decomposition based on Rh catalysts has made significant progress. Their unique catalytic performance and stability provide a new research direction for the efficient conversion of NO_x.⁴¹

DND is thermodynamically spontaneous (ΔG_{100 °C} = −86.6 kJ mol⁻¹). However, the strong binding energy of the N–O bond (about 630.6 kJ mol⁻¹) leads to a higher activation energy (about 335 kJ mol⁻¹), making the key break become the speed limiting step.⁴² Secondly, the strong adsorption of by-product oxygen on the catalyst surface will poison the active sites and significantly inhibit the reaction process. In addition, the presence of impurities such as SO₂ and steam under industrial conditions, as well as the high-temperature operating environment required for the reaction, poses severe challenges to the stability of the catalyst, which together restrict the industrialization process of DND technology.

In the DND reaction, it is widely accepted that the intermediate product is N₂O, which subsequently decomposes, ultimately yielding N₂ and O₂.⁴³ The reaction pathway after the formation of N₂O was investigated in detail by the researchers. Most studies believe that N₂O is adsorbed on the active site of the catalyst and decomposed into gaseous N₂ and adsorbed atomic O, then another N₂O molecule is adsorbed on this adsorption site and decomposed into gaseous N₂ and adsorbed atomic O, and then the two adsorbed atomic O combine to form an O₂ molecule and desorb.^44,45 The formula is as follows:

2NO → N₂O_(ads.) (18)

N₂O_(ads.) → O_(ads.) + N_2(g) (19)

O_(ads.) + O_(ads.) → O_2(g) (20)

To gain a deeper and intuitive understanding of the DND reaction, Fig. 4 shows the decomposition mechanism of NO on the surface of C-type cubic rare earth oxide. NO reacts with the surface active sites of rare earth oxides to form monodentate nitrates (the structure is O–N–O⁻ bonded to the surface). Monodentate nitrates can decompose to form N-containing species, which interact with surface oxygen species to further generate N₂O, while the structure of surface active sites changes (such as (O⁻)(O)(O⁻) to (O⁻)( ⁻)(O⁻)(O⁻)). Monodentate nitrates can also be converted into nitrosyl groups. Nitrosyl species react with NO and O₂ to produce N₂, and return the surface sites to the initial state. The results show that the direct decomposition of NO is carried out by the reaction of adsorbed NO_x species with gaseous NO. In this mechanism, the surface basic center plays an important role in the adsorption of NO, and the anion vacancy is necessary for the formation of nitrosyl. Therefore, we can increase the indicated basic sites and anion vacancies by adding cations (such as Ba²⁺), which will significantly improve the catalytic activity.

Fig. 4 Overall reaction mechanism for direct NO decomposition on a C-type cubic rare earth oxide-based catalyst⁴⁶ (Copyright 2013, RSC).

2.1.3 Other reactions. In addition to direct decomposition and selective reduction, other methods can also be used to realize exhaust gas treatment in actual production, such as non-selective catalytic reduction, storage reduction, etc.

Non selective catalytic reduction (NSCR) means that the reductant reacts with O₂ in the atmosphere first, and then reacts with NO_x and reduces it to N₂.⁴⁷ Compared with the direct decomposition method and SCR technology, the disadvantage of this method is that the amount of reductant is very large, and a strong thermal effect of the catalytic layer is caused, resulting in the reduction of catalytic stability, activity and lifespan. Three-way catalysts (TWCs) widely used in automobiles are typical NSCR catalysts, which can catalyze NO_x, CO and HCs in automobile exhaust to produce N₂, CO₂ and H₂O, respectively. Fig. 5 shows the reaction mechanism of a Rh-based catalyst in a three-way catalyst (TWC). In a high temperature oxidizing atmosphere, the Rh metal is oxidized and interacts with the support. During this process, the Rh metal is irreversibly sintered, resulting in permanent deactivation of the catalyst. In a reducing atmosphere, reducing substances (such as H₂ and CO) restore the oxidized Rh, reverse the “oxidation metal–support interaction”, and regenerate the catalyst. Fig. 5 shows that in actual work, Rh-based catalysts should be kept in a reducing atmosphere as much as possible to realize sustainable regeneration of the catalyst, to reduce the negative impact caused by catalyst deactivation.

Fig. 5 Reaction mechanism of a Rh-based catalyst in a TWC⁴⁹ (Copyright 2015, MDPI).

NO_x storage-reduction (NSR) was proposed in the 1990s. NO_x-NSR technology mainly includes five stages:⁴⁸ (1) NO is oxidized to NO₂. Because the adsorption capacity of NO is lower than that of NO₂, this reaction is very important in the NSR process; (2) NO and NO₂ interact with BaO to store NO_x in the form of barium nitrite and barium nitrate; (3) the reaction atmosphere turns to rich combustion, and reductants such as CO, HCs and H₂ are introduced to consume excess O₂; (4) the decomposition of barium nitrite or barium nitrate releases NO and NO₂; (5) NO and NO₂ react with reducing agents such as CO, HCs and H₂ to form N₂.

2.2 Reaction mechanism

2.2.1 Langmuir–Hinshelwood (L–H) mechanism. The Langmuir–Hinshelwood (L–H) mechanism, proposed by Irving Langmuir and Cyril Hinshelwood, describes heterogeneous catalytic reactions involving co-adsorbed reactants, which are widely applied in gas–solid systems (e.g., CO oxidation, NO_x reduction, and VOC treatment). In NH₃-SCR, the specific mechanism starts from the adsorption of NH₃, and the physically adsorbed NO can be oxidized by M in the form of nitrite and nitrate through reactions (23) and (24), respectively. Through reactions (25) and (26), these substances react with adsorbed NH₃ to produce NH₄NO₂ and NH₄NO₃, which are then decomposed into N₂ and N₂O, respectively. Finally, the reduced Mⁿ⁻¹ can be rapidly regenerated by O₂.⁵⁰ The specific reaction is as follows:

NH_3(g) → NH_3(ads.) (21)

NO_(g) → NO_(ads.) (22)

Mⁿ⁺ = O + NO_(ads.) → M⁽ⁿ⁻¹⁾⁺–O–NO (23)

M⁽ⁿ⁻¹⁾⁺–O–NO + NH_3(ads.) → M⁽ⁿ⁻¹⁾⁺–O–NO–NH₃ → M⁽ⁿ⁻¹⁾⁺–O–H + N₂ + H₂O (25)

M⁽ⁿ⁻¹⁾⁺–O–NO₂ + NH_3(ads.) → M⁽ⁿ⁻¹⁾⁺–O–NO₂–NH₃ → M⁽ⁿ⁻¹⁾⁺–O–H + N₂O + H₂O (26)

By adjusting the reaction conditions, researchers found that in the CO-SCR reaction, when the temperature rises to 300–500 °C, the reaction also follows the L–H mechanism. Fig. 6 (ref. 51) shows the Langmuir Hinshelwood (L–H) mechanism of CO selective catalytic reduction of NO (CO-SCR). CO and NO molecules are first adsorbed on the catalyst surface to form adsorbed CO_x species and NO_x species. Then, these species adsorbed on the catalyst surface interact and react to form N₂ and CO₂, and the product desorbs from the catalyst surface to complete the reaction cycle. This figure intuitively shows the core feature of the L–H mechanism, that is, the substances involved in the reaction need to be adsorbed to the active site first and react in the form of adsorption state to complete the degradation of waste gas. This makes the active sites that determine the adsorption performance of the catalyst become the top priority.

Fig. 6 L–H mechanism in CO-SCR⁵¹ (Copyright 2023, ELSEVIER).

The basic principle of the L–H mechanism in H₂-SCR is that NO is decomposed and converted into nitrogen-containing products by reaction. The specific reaction is as follows (M* indicates the active site on the precious metal):³⁵

M* + NO → NO_(ads.) (28)

H₂ + 2M* → 2H_(ads.) (29)

O₂ + 2M* → 2O_(ads.) (30)

NO_(ads.) + 2M* → N_(ads.) + O_(ads.) (31)

N_(ads.) + N_(ads.) → N₂ + 2M* (32)

N_(ads.) + NO_(ads.) → N₂O + 2M* (33)

O_(ads.) + H_(ads.) → OH_(ads.) + 2M* (34)

OH_(ads.) + H_(ads.) → H₂O + 2M* (35)

NO_(ads.) + H_(ads.) → N_(ads.) + OH_(ads.) (36)

NO_(ads.) + NO → N₂O + O_(ads.) (37)

In this mechanism, H₂ plays a role of consuming adsorbed oxygen on the catalyst surface and releasing more M* active vacancies, which indirectly promotes the adsorption and dissociation of NO.

2.2.2 Eley–Rideal (E–R) mechanism. The Eley–Rideal (E–R) mechanism is a classical model describing heterogeneous catalytic reactions in which one reactant is adsorbed on the catalyst surface, and the other reacts directly with adsorbed species in the gas phase or weak adsorption state. Its core feature is that one reactant is chemically adsorbed on the catalyst surface, and the other reactant directly participates in the reaction in the gas phase or physical adsorption state, which is suitable for some hydrogenation, oxidation or reduction reactions. The E–R mechanism in NH₃-SCR is that the adsorbed NH₃ is activated to NH₂ (reaction (2)) and then oxidized to NH by metal ions. NH₂ and NH react with NO to produce N₂ and N₂O, respectively.⁵⁰ The specific reactions are as follows:

NH_3(g) → NH_3(ads.) (38)

NH_3(ads.) + Mⁿ⁺ = O → NH_2(ads.) + M⁽ⁿ⁻¹⁾⁺–O–H (39)

NH_2(ads.) + Mⁿ⁺ = O → NH_(ads.) + M⁽ⁿ⁻¹⁾⁺–O–H (40)

NH_2(ads.) + NO_(g) → N₂ + H₂O (41)

NH_(ads.) + NO_(g) → N₂O + H⁺ (42)

As mentioned above, CO-SCR follows the L–H mechanism at higher temperatures. When the temperature is lower, the E–R mechanism will become dominant. As shown in Fig. 7, below 200 °C, CO is adsorbed on the surface of the catalyst in a gaseous state, and NO molecules are preferentially adsorbed on the active centers below 200 °C, generating various nitrate species in the low oxidation state. Then they react with adsorbed CO under the action of active centers (Mⁿ⁺). CO is oxidized to CO₂, NO and O dissociated from nitrate species form O₂ and N₂, and the final reaction is completed. The figure shows the core of the E–R mechanism, that is, the co-participating NO and CO in the reaction directly reacts in the gas phase, rather than reacting with the active center first. Compared with L–H, the adsorption process of reactants is reduced, making the reaction simpler and faster, but at the same time, the impurities in the atmosphere may have a certain impact on the reaction. The specific reaction steps are as follows:⁵²

Mⁿ⁺ + NO → M⁽ⁿ⁺¹⁾⁺ + N₂O (43)

M⁽ⁿ⁺¹⁾⁺ + NO + CO → Mⁿ⁺ + CO₂ + N + O (44)

NO → N + O (45)

CO + O → CO₂ (46)

NO + N → N₂O (47)

N₂O → O + N₂ (48)

N + N → N₂ (49)

Fig. 7 The proposed CO-SCR reaction mechanism on a Ce_x-OMS-2 catalyst⁵² (Copyright 2023, ELSEVIER).

As mentioned above, there are three mechanisms in H₂-SCR: the NO adsorption/desorption mechanism; the redox mechanism of NO; a bifunctional mechanism.⁵³ The NO adsorption/desorption mechanism is also known as the L–H mechanism, while the NO oxidation–reduction mechanism is the same as the E–R mechanism. NO is oxidized to nitrogen-containing intermediates (such as NO₂, NH₃, and NO³⁻) under the action of O₂ and then reduced to N₂ and H₂O under the action of H₂. That is, a reactant is adsorbed on the catalyst surface and reacts directly with gaseous reactants.

2.2.3 Mars–van Krevelen mechanism. The Mars–van Krevelen (MvK) mechanism is an important model to describe the oxidation–reduction reaction in heterogeneous catalysis, especially for catalytic processes involving metal oxide catalysts. The core feature of this mechanism is that lattice oxygen directly participates in the reaction, and the catalyst achieves catalytic activity through periodic redox cycles. This mechanism is often used to describe the selective oxidation of hydrocarbons, such as the oxidation of propylene to acrolein. It can also be used in the CO-SCR catalytic reaction, but it usually cooperates with the L–H or E–R mechanism. As shown in Fig. 8, in an Fe-based catalyst, Fe₂O₃ supplies lattice oxygen according to the MvK mechanism. The reactant CO is first adsorbed onto the catalyst surface and then reacts with lattice oxygen to generate CO₂. CO₂ is desorbed from the catalyst surface to form an oxygen vacancy (O_V) and then promotes the reaction of NO to N₂O. The oxygen vacancy is refilled, and then forms CO₂ through CO to form an oxygen vacancy again, which promotes the transformation from N₂O to N₂, and the oxygen vacancy is transformed into lattice oxygen, a catalytic loop of oxygen vacancy generation–filling can be produced on the α-Fe₂O₃ (1 1 0) surface in CO-SCR, and the continuous catalytic process is realized by charge transfer between reactive species.⁵⁴ Although this mechanism faces the challenge of lattice stability, its application in selective oxidation and environmental catalysis is still irreplaceable due to the efficient reduction of NO_x and the continuous regeneration of the catalyst.

Fig. 8 Schematic mechanism of the standard CO-SCR reaction over the α-Fe₂O₃ (1 1 0) surface⁵⁴ (Copyright 2023, ELSEVIER).

3. Common deactivation and mechanism of Rh-based catalysts

In complex gas environments, catalysts often encounter multiple challenges, including the mixing of various gases and temperature fluctuations. These complex conditions may trigger unnecessary reactions, leading to catalyst deactivation. For Rh-based catalysts in exhaust gas treatment, deactivation results from the synergistic effects of multiple factors. Based on the involved reactions and external factors, deactivation is classified into chemical deactivation and physical deactivation.

3.1 Chemical deactivation and mechanism

Chemical deactivation is one of the main factors leading to the performance degradation of Rh-based catalysts. Chemical deactivation refers to the phenomenon whereby during the catalytic reaction process, chemical reactions with harmful impurity elements in the flue gas significantly reduce or even eliminate the catalyst's activity, selectivity, or stability. Studies indicate that the chemical deactivation mechanism primarily manifests through the adsorption of poisons onto the catalyst or chemical reactions between poisons and the catalyst during the catalytic process. These processes destroy active components or alter chemical properties, reducing activity.⁵⁵ The poisoning of Rh-based catalysts can be categorized into sulfur poisoning, NO_x poisoning, and H₂O poisoning. Different types of chemical deactivation and their mechanisms, commonly observed in Rh-based catalysts, will be discussed in detail below.

3.1.1 Deactivation and mechanism of sulfur poisoning. The deactivation of Rh-based catalysts by sulfur poisoning primarily involves a competitive adsorption mechanism. In this process, sulfur compounds compete with reactant molecules for the active sites on Rh. When sulfur exhibits a stronger adsorption affinity, it occupies these sites preferentially, thereby deactivating the catalyst. Noble metals (e.g., Rh and Pt) readily interact with sulfide sp-hybrid orbitals due to their electronic structures, forming stable sulfur bonds that cause poisoning.⁵⁶ Second, adsorbed sulfide reacts with surface lattice oxygen or adsorbed oxygen to form sulfate species, resulting in active site occupation. Zhang et al.²⁵ studied sulfur poisoning of a Rh/ZSM-5 catalyst. Compared to conditions without SO₂, when the reaction atmosphere contains 1 ppm SO₂, the catalyst activity at 400 °C decreases by approximately 50%, and the operating window shifts entirely to higher temperatures. This indicates that catalyst activity is lost following SO₂ introduction. Subsequently, the researchers quantitatively described SO₂ coverage on the catalyst surface using the Temkin adsorption isotherm and attributed catalyst deactivation to SO₂ adsorption occupation on Rh₂O₃ active sites. Ocsachoque et al.⁵⁷ studied sulfur poisoning of a Rh/CeO₂ catalyst. When SO₂ was added to the reaction atmosphere, the catalytic performance of Rh/CeO₂ samples showed residual activity equal to 0.45 of the initial activity. Raman spectroscopy analysis revealed that introduced H₂S reacts with the metal to produce M–S species, which occupy active sites and cause catalyst poisoning, and further generates sulfate, aggravating catalyst poisoning and deactivation. Density functional theory (DFT) calculations show that sulfur adsorption alters the bond lengths between related atoms on the catalyst surface, significantly reducing the catalyst's adsorption performance. Wang et al.⁵⁸ reviewed research by numerous investigators on catalyst sulfur poisoning, summarized the detailed process of SO₂ adsorption, migration, and conversion on the catalyst, and systematically revealed the complete process of catalyst sulfur poisoning leading to deactivation. Fig. 9 illustrates the reaction process of sulfide reacting with lattice oxygen on the metal surface to form sulfate. First, sulfide forms initial monodentate adsorption with lattice oxygen through strong chemical adsorption and subsequently reacts with gaseous oxygen to form bidentate sulfate. With increasing reaction temperature and sulfide concentration, the initially formed bidentate sulfate gradually transforms into a more thermodynamically stable structure. When H₂O participates in the reaction, the monodentate species transforms into bridged bidentate sulfate, then into a chelated bidentate structure, and further into tridentate sulfate. This structural evolution of sulfate enhances the stability of surface sulfides and further aggravates catalyst deactivation. Fig. 9 shows that SO₂ primarily leads to catalyst deactivation by forming various types of sulfates that occupy the active sites. The presence of H₂O in the atmosphere can also impair the sulfur storage capacity of the support, thereby accelerating the deactivation of the precious metal active sites.

Fig. 9 Schematic diagram of the sulfate formation mechanism in (a) anhydrous and (b) water environments⁵⁸ (Copyright 2024, ELSEVIER).

The essence of sulfur poisoning is that sulfur species occupy or destroy active sites on the catalyst surface through irreversible chemical adsorption or reaction, leading to the loss of catalytic functionality. It is a dynamic and multi-stage chemical conversion process. It starts from the competitive adsorption of sulfide, while the subsequent irreversible sulfation reaction and structural evolution are also important factors leading to catalyst deactivation. Therefore, the development of sulfur-resistant catalysts or the study of regeneration strategies must focus on inhibiting the adsorption of sulfur substances and the formation of sulfate, particularly preventing its evolution to the most thermodynamically stable high coordination/polymerization state structure. Understanding and blocking this evolution process are the central challenges in improving the sulfur resistance of catalysts.

3.1.2 Deactivation and mechanism of NO_x poisoning. For Rh-based catalysts where NO_x is a reactant, it does not merely participate in the reaction but can also induce poisoning. This form of deactivation stems from the occupation and passivation of active sites by reaction by-products and strongly adsorbed oxygen species (O_ad) generated during the NO_x conversion process.^59,60 Wang et al.⁶¹ studied Rh/CeO₂ performance for selective NO reduction. Performance tests showed that when the catalyst was re-oxidized, NO conversion decreased sharply while NO₂ by-products appeared simultaneously. Mass spectrometry (MS) and Fourier transform infrared spectroscopy (FT-IR) studies revealed that after reaction periods, NO and O₂ compete for adsorption on oxygen vacancies. When oxygen vacancies were filled, catalyst deactivation occurred, with this competitive effect being particularly significant at high temperatures and high space velocities. This indicates that the primary factor in NO_x poisoning and catalyst deactivation is surface active site occupation. Li and Hadjiivanov et al.^59,62 confirmed through in situ DRIFTS studies that O_ad species generated by NO dissociation on Rh-based catalysts exhibit strong adsorption characteristics. If these species cannot be desorbed and reconstituted promptly, continuous passivation of active sites occurs, weakening adsorption capacity and causing NO_x poisoning. Bae et al.⁶³ systematically studied Rh/PrO_x deactivation mechanisms using DRIFTS, NO-temperature programmed desorption (NO-TPD), near edge X-ray absorption fine structure (NEXAFS), and other characterization techniques. Experimental results showed that preferential active site occupation by adsorbed oxygen caused catalyst deactivation. Fig. 10 illustrates the mechanism of oxygen occupying catalyst active sites during NO degradation. First, NO adsorbs on catalyst active sites and decomposes into adsorbed N and adsorbed O. The combination of adsorbed N and O generates N₂ and desorbs, while undissociated adsorbed oxygen occupies metal active sites, hinders NO adsorption/reaction, and oxidizes metallic Rh to inert oxidized Rh species, reducing reduction activity. The combination of strong binding of O with facile O dissociation kinetics triggers the oxygen poisoning on the metallic phase. According to the characteristics of exothermic N–O bond cleavage, competitive generation of N₂ and metal phase poisoning, the deactivation process is attributed to the occupation of Rh metal active sites by oxygen atoms.

Fig. 10 Schematic diagram of oxygen occupied active sites⁶³ (Copyright 2023, ELSEVIER).

The essence of NO_x poisoning of Rh-based catalysts is that adsorbed oxygen (O_ad) reduces the number of active sites and weakens their reduction ability through the dual mechanism of occupying active sites and oxidizing Rh's metallic state. Among these, the strong adsorption characteristics of O_ad and its induced oxidation of Rh metal are the main reasons for deactivation, while competitive adsorption by nitrogen-containing intermediates (NO⁻/NO₃⁻) further aggravates the activity decline. Therefore, inhibiting NO_x poisoning on Rh-based catalysts should consider optimization of O_ad desorption kinetics and anti-oxidation surface engineering.

3.1.3 Deactivation and mechanism of H₂O poisoning. The deactivating effect of H₂O vapor, though relatively rare, is a non-negligible phenomenon in Rh-based catalysis. Its mechanism operates through the competitive adsorption of H₂O on Rh active sites or the acidic sites of the support, thereby inducing surface hydroxylation and resulting in site blocking.⁶⁴ In studies by Dai and Yang et al.,^65,661H MAS NMR (magic angle rotation nuclear magnetic resonance technology) revealed that the characteristic peak (4.2 ppm) representing free Brønsted acid sites disappeared in the deactivated catalyst spectra, while a broad peak appeared at 5–9 ppm, corresponding to hydrogen-bonded complexes formed between H₂O molecules and acid sites. Additionally, in situ DRIFTS data demonstrated that under humid atmospheres, strong broad peaks (3200–3600 cm⁻¹) appeared on catalyst surfaces, representing hydrogen bonded OH groups, while the signal for free acid sites (3645 cm⁻¹) weakened. These data indicated that competitive adsorption of H₂O molecules reduced available Brønsted acid sites, formed hydrogen-bonded complexes, blocked active sites, and led to catalyst deactivation. Additionally, Abdullah et al.⁶⁷ used specific surface area and pore structure analyses to conclude that H₂O molecule addition forms an adsorption layer or a liquid film at catalyst pore necks, hindering reactant diffusion/mass transfer and inhibiting harmful gas degradation, particularly in catalysts with small pores. Following steam introduction, hydroxyl species (–OH or OH radicals) can form on catalyst surfaces.⁶⁸ Persson et al.⁶⁹ studied H₂O's effect on noble metal catalyst performance. Experimental results showed that hydroxyl species negatively affected catalyst activity, particularly for VOCs where C–H bond activation is the decisive reaction step; surface hydroxylation slowed down this step, causing the activity decline. Li et al.⁷⁰ studied H₂O's effect on N₂O catalytic decomposition over Rh catalysts, finding that after adding 5% H₂O to the atmosphere, N₂O conversion over fresh Rh/CeO₂ decreased from 75% to 8%. DRIFTS identified adsorption species on catalyst surfaces. Experimental data showed that surface –OH inhibited N₂O adsorption, and stable surface –OH occupied and blocked oxygen vacancies, inhibiting the reaction. Comprehensive data indicated that catalyst deactivation primarily resulted from H₂O poisoning due to surface hydroxylation following H₂O addition. Zhang et al.⁷¹ studied VOC catalytic oxidation mechanisms in H₂O and revealed the complete catalyst surface hydroxylation process via in situ infrared spectroscopy and other experiments. Fig. 11 illustrates the mechanism of H₂O conversion into hydroxyl groups on catalyst surfaces. First, gaseous or liquid H₂O molecules adsorb onto Rh surface-active sites via hydrogen bonding. Then, H₂O dissociates into adsorbed hydrogen and hydroxyl under acid site catalysis. However, hydroxyl desorption is difficult, leading to continuous hydroxyl accumulation. Hydroxyl species passivate the catalyst's electronic structure, ultimately reducing activity. Overall, the conversion of H₂O to hydroxyl on the surface of the catalyst to occupy the active site and hinder the reaction is the key step to the deactivation of the catalyst. The H₂O resistance of catalysts has also become an important indicator to measure the catalytic performance.

Fig. 11 Conversion of H₂O to hydroxyl on the catalyst surface⁷¹ (Copyright 2025, ELSEVIER).

The H₂O poisoning of Rh-based catalysts involves deactivation primarily through competitive adsorption and surface hydroxylation by H₂O molecules. Its core lies in the dual role of hydroxyl species: physical occupation blocking active sites inhibits reactant adsorption, while chemical modification of active substances' electronic structures inhibits catalytic ability. This process is especially severe in microporous catalysts and under low-temperature, high-humidity conditions. Inhibition strategies can achieve physical shielding of H₂O molecules through hydrophobic structure construction and enable dynamic hydroxyl removal via support design, thereby suppressing the H₂O poisoning phenomenon in Rh-based catalysts.

3.1.4 Other chemical deactivation mechanisms. Among the deactivation mechanisms of Rh-based catalysts, in addition to the above-mentioned three deactivation mechanisms caused by chemical poisoning, the chemical changes of Rh and the interaction between Rh and the support are also noteworthy. They can be roughly divided into two deactivation modes: (1) the direct deactivation of Rh, which is mainly due to the chemical state transition of Rh in a high temperature oxidation atmosphere (such as Rh⁰ → Rh³⁺).⁷² The conversion of the active metal Rh⁰ to Rh³⁺ causes the reorganization of the surface electronic structure, and the metal Rh⁰ with higher electron density has a stronger ability to dissociate NO than the electron deficient Rh³⁺. The stronger NO dissociation ability of metal Rh⁰ is due to its electronic feedback to the 2π* antibonding orbital of NO, which reduces the activation barrier of the broken N–O bond.^73,74 Haneda et al.⁷⁵ studied the effect of the Rh valence on the TWC performance of the catalyst. In situ infrared spectroscopy and XANES spectroscopy data showed that if the Rh species in the catalyst were in a stable oxidation state, its relatively high OSC value would lead to lower TWC performance. (2) The deactivation caused by the interaction between Rh and the support (such a support is usually Al₂O₃). This interaction will make Rh diffuse into Al₂O₃ to form RhAlO_x. The formation of this new material will first cause the active metal to be occupied and unable to participate in the catalytic process, but this is reversible. The reduction of the sample can restore the activity of noble metal Rh. In addition, the formation of RhAlO_x will also lead to the phase transition of Al₂O₃, which will have an irreversible effect on the catalyst (the support phase transition caused by the composite phase will be described in the second part of section 3.2.2). Machida et al.²⁴ studied the thermal behavior of Rh loaded Al₂O₃ at high temperature and under different atmospheres through X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and extended X-ray absorption fine structure (EXAFS). Fig. 12 shows the difference between Rh-based catalyst and Pd-based catalyst aging under the same conditions. Deactivation of Pd-based catalysts is mainly the sintering of precious metals, while that of Rh-based catalysts is dominated by the interaction between precious metals and supports. Therefore, when facing the high temperature deactivation of Rh-based catalysts, in addition to the sintering of precious metals, the interaction between metals and supports should also be considered.

Fig. 12 Surface changes of Rh/Al₂O₃ and Pd/Al₂O₃ in air and H₂ (ref. 24) (Copyright 2019, ACS).

The deactivation of Rh-based catalysts in a high temperature oxidation environment is due to the chemical evolution of noble metals and supports: the transition from Rh⁰ to the Rh³⁺ oxidized state, and the solid-state reaction of oxidized Rh and supports to produce the RhAlO_x substance, which covers the active interface. The formation of RhAlO_x is the main cause of catalyst deactivation. When studying the inhibition strategy, it is necessary to consider blocking the Rh–O–Al bond generation path and maintaining the dynamic balance of Rh⁰ by designing the antioxidant interface or stabilizing the support structure.

3.2 Physical deactivation

Physical deactivation refers to the phenomenon that the activity, selectivity and long-term operation stability of Rh-based catalysts significantly decrease due to the microstructure evolution of catalyst components driven by heat during high-temperature reactions. Physical deactivation includes active component sintering, support sintering, phase transformation and other deactivation mechanisms. Sintering is the most common deactivation phenomenon in Rh-based catalysts. Additionally, although support phase transformation accounts for a relatively minor portion of catalyst deactivation mechanisms, its impact on catalyst performance cannot be ignored. The common physical deactivation mechanisms of Rh-based catalysts are discussed in detail below.

3.2.1 Rh sintering deactivation and mechanism. In Rh-based catalyst systems, Rh serves as an important active site, and its high-temperature thermodynamic behavior has consistently been a research focus. Rh sintering refers to the phenomenon whereby highly dispersed Rh metal nanoparticles aggregate at elevated temperatures or under specific reaction conditions, resulting in increased particle size, decreased specific surface area, reduced active site density, and diminished catalytic performance. Key factors include temperature (high temperature significantly accelerates atomic diffusion and particle migration) and support properties (weak metal support interaction, such as SiO₂ loaded Rh, cannot anchor particles and promote sintering). Sintering primarily follows particle migration and coalescence (PMC) and Ostwald ripening (OR) mechanisms. As shown in Fig. 13, under the PMC mechanism, Rh metal grains migrate and aggregate along the support surface, followed by collision and merging of adjacent grains, ultimately forming larger spherical particles; in the OR mechanism, smaller particles with higher curvature dissolve, and the dissolved species diffuse through the medium to deposit onto larger particles with lower solubility, resulting in large-particle growth (maturation), ultimately increasing Rh particle size and decreasing active specific surface area.^76,77 The sintering mechanism shown in Fig. 13 is helpful to better determine how the particles grow in the actual sintering scheme and control the size of metal particles through the particle growth mechanism in subsequent studies. Nakayama et al.⁷⁸ observed the atomic-scale high-temperature sintering behavior of Rh nanoparticles on various oxide supports (ZrO₂, Y-ZrO₂, CeO₂, and γ-Al₂O₃) using in situ transmission electron microscopy. Rh particle sintering initiated above 900 °C, with significant agglomeration occurring above this temperature. Fig. 14 shows the transmission electron microscopy (TEM) images and energy dispersive X-ray spectroscopy (EDS) spectra of the Rh/Y-ZrO₂ catalyst at different aging temperatures. It can be observed from the figure that with the increase of temperature, Rh begins to agglomerate obviously, which indicates that Rh will be sintered obviously at 900 °C or above. Varga et al.⁷⁹ examined Rh particle size changes post-heat-treatment using high-resolution electron microscopy. Results showed that the average Rh particle size increased from 2.3 nm to 3.3 nm after 500 °C heat treatment, with particles exceeding 5 nm appearing in high-loading samples. Goula et al.⁸⁰ studied the thermal sintering behavior of Rh particles. TEM results indicated that Rh on γ-Al₂O₃ could grow significantly (an increase of about 117%) under oxidation treatment at 850 °C. These studies confirm that treatment temperatures beyond specific thresholds induce Rh nanoparticle sintering, potentially causing a decline in catalyst performance.

Fig. 13 Schematic diagram of the sintering mechanism⁸⁰ (Copyright 2019, MDPI).

Fig. 14 TEM images and EDS mapping of Rh in Rh/Y-ZrO₂ under vacuum conditions, the temperatures are (a) 300 °C, (b) 900 °C, (c) 950 °C, (d) 1000 °C and (e) 1050 °C, respectively⁷⁸ (Copyright 2021, MDPI).

The deactivation of Rh-based catalysts due to Rh sintering primarily results from PMC and OR mechanisms. Ostwald ripening dissolves small particles in low-temperature regions, while particle migration and aggregation dominate coarsening in medium-high temperature regions; deactivation occurs through coupling of both mechanisms above 900 °C. Support interfacial properties (solubility, diffusion coefficient, and interface energy) constitute core factors influencing Rh sintering. The inhibition strategy involves designing anti-sintering structures that anchor Rh particles (inhibiting migration) and optimize particle size to withstand harsh high-temperature catalyst operating conditions.

3.2.2 Support deactivation and mechanism.
(1) Support sintering. Support sintering refers to structural collapse, pore closure, and related phenomena in catalyst support materials (e.g., Al₂O₃, SiO₂, and CeO₂) at elevated temperatures, leading to reduced specific surface area and pore volume, thereby weakening active component dispersion and mass transfer efficiency. Key factors include the support type and hydrothermal environment. The particle migration and coalescence (PMC) and Ostwald ripening (OR) mechanisms mirror those observed in active material sintering. Lee et al.⁸¹ studied the denitration performance of a zeolite catalyst. BET data showed that after hydrothermal aging at 700 °C, the total pore volume decreased from 0.5373 cm³ g⁻¹ to 0.4749 cm³ g⁻¹, and the micropore volume decreased from 0.1453 cm³ g⁻¹ to 0.0681 cm³ g⁻¹. Support sintering reduced NO_x conversion from 60% to 20% at 450 °C. Additionally, support encapsulation phenomena constitute another form of sintering. Machida et al.⁸² reported that the specific surface area of a Rh/ZrO₂–CeO₂ ternary catalyst decreased from 56.8 m² g⁻¹ to 17.1 m² g⁻¹ and the OSC (oxygen storage capacity) decreased from 0.99 g to 0.10 g after thermal aging in a cycle from stoichiometry to oxidation to reduction (SLR cycle). Performance comparisons across atmospheres (lean burn, rich burn, and dynamic SLR cycles) revealed that dynamic SLR conditions caused the most severe adsorption degradation (82% NO adsorption reduction). High-resolution transmission electron microscopy (HRTEM) and X-ray spectroscopy analyses (Fig. 15a–c) showed complete encapsulation of Rh nanoparticles by the CeO₂–ZrO₂ (CZ) support during SLR cycling, with Rh particles exhibiting a 0.26 nm lattice fringe spacing attributed to Zr_0.8Ce_0.2O₂. Fig. 15d illustrates the process through which the support progressively encapsulates and buries the surface-active species. Under high-temperature SLR cycling, the Rh/ZC interface undergoes rapid, repetitive oxygen release–storage processes driven by cyclically changing conditions. This activates migration of the ZC components, thereby accelerating the overcoating of Rh nanoparticles and ultimately leading to the formation of a physical blocking encapsulation. In addition to the above sintering and encapsulation of the support caused by high temperature, the interaction between Rh and the support itself may also cause the degradation of the support structure. Alikin et al.⁸³ systematically studied the effect of Rh loading (0.01–1 wt%) on the structural stability of a variety of cerium-based materials (CeO₂, CeZrO₂, and ZrCeYLaO₂) after high temperature aging (1000–1100 °C). It is found that the introduction of Rh will accelerate the collapse of the support structure, which shows the decrease of specific surface area and the increase of average pore size. It is particularly noteworthy that even without phase transition (no new phase was detected by XRD), the formation of Rh dense and cerium rich particles was observed by scanning electron microscopy (SEM) and density measurement, which indicates that the local structure of the support is densified. This densification effect becomes significant with the increase of Rh loading to about 0.1 wt%, and its negative effect follows the order CeO₂ > CeZrO₂ > ZrCeYLaO₂. This indicates that in Rh-based catalysts, the active metal itself may also be an important factor driving the physical structure degradation and deactivation of the support. Support structural deterioration (e.g., reduced specific surface area, collapsed pore structure) and active component encapsulation constitute the main deactivation pathways from support sintering. This highlights the necessity of considering high-temperature support physical properties during Rh catalyst preparation. For inhibition strategies, supports with high thermal stability must be designed to maintain functionality under high-temperature operation.

Fig. 15 HRTEM images of Rh/ZC under (a) rich burn conditions (R) and (b) stoichiometric lean burn (SLR) cycle conditions after engine aging at 1000 °C for 40 hours; (c) X-ray spectra obtained from the red and blue regions of image (b); (d) schematic diagram of the possible Rh encapsulation modes induced by the burial by the ZC support during the grain growth and the collapse of the porous support structure and activated migration of ZC components during repeating oxygen storage/release under dynamic SLR cycle conditions⁸² (Copyright 2023, ACS).

(2) Support phase transition. Deactivation caused by support phase change refers to irreversible crystal structure transformations of catalyst support materials (e.g., oxides and molecular sieves) due to heat, chemical environment changes, or mechanical stress during catalytic reactions or regeneration, resulting in significant degradation of physicochemical properties (e.g., specific surface area, pore structure, and surface acidity/alkalinity). This weakness supports stability for active components or directly destroys active sites, ultimately leading to a decline in the overall catalyst performance. The most common support phase transition in Rh-based catalysts occurs in systems using Al₂O₃ as the support. Research shows phase transition from γ-Al₂O₃ to θ-Al₂O₃ (predominant) and α-Al₂O₃ (minor) occurs at high temperatures (>1200 °C), as shown in Fig. 16.⁸⁴ Stoyanovskii et al.⁸⁵ also studied the structure of a Rh/Al₂O₃ catalyst after calcination at high temperature, using laser-induced luminescence and X-ray diffraction techniques. The experimental results show that when the Rh/Al₂O₃ catalyst is calcined in an oxidizing atmosphere above 1000 °C, Rh³⁺ ions can promote the phase transition from γ-Al₂O₃ to α-Al₂O₃. Specifically, Rh³⁺ ions dissolve in the low-temperature phase of alumina and are captured and stabilized by the newly formed α-Al₂O₃ structure during high-temperature heat treatment. This process not only leads to the doping of Rh³⁺ ions into the corundum phase but also acts as an “inducer” or “promoter” at the early stage of phase transition, promoting the formation of a new phase, and finally causing irreversible deactivation of the catalyst. The formation of the new phase not only reduces the specific surface area of the support, but also reduces the dispersion of Rh particles, making the density of strong acid centers increase from low to high, and leading to the decline of activity.^86,87 Additionally, phase transition alters the support physical and chemical properties. Rh–support interactions vary across different supports. Results indicate that Rh and γ-Al₂O₃ form strong interactions at heat treatment temperatures above 600 °C in air, while α-Al₂O₃ requires temperatures up to 900 °C,⁸⁸ leading to RhAlO_x species formation. Strong Rh–support interactions cause significant active site loss, resulting in a decline of catalyst activity. Phase transition deactivation in Rh-based catalysts stems from structural degradation triggered by crystal reconstruction. This degradation involves reduced specific surface area, distorted strong acid sites, and Rh–O–Al bonding mechanisms. To suppress support phase transition, it is essential to move beyond traditional Rh-based catalyst support materials, balance performance and stability, and achieve coordinated regulation of support structural stability and catalytic activity.

Fig. 16 XRD patterns over fresh and aged Rh/γ- and θ-Al₂O₃ catalysts⁸⁴ (Copyright 2021, MDPI).

3.2.3 Other physical deactivation mechanisms. In addition to the aforementioned common physical deactivation mechanisms, catalyst performance is also degraded by thermal stress at elevated temperatures. Thermal stress refers to internal stress within catalysts at high temperatures or under rapid temperature changes, resulting from differences in thermal expansion coefficients or internal temperature gradients. Thermal stress impacts catalyst performance primarily through: structural damage, where interfacial cracks between Rh particles and the support may cause particle detachment,⁸⁹ and mechanical strength reduction, decreasing catalyst robustness and increasing vulnerability to physical damage during operation.⁹⁰ Although not a common deactivation factor, thermal stress cannot be ignored in catalyst system design, as it critically impacts high-temperature thermal stability and adaptability to complex dynamic conditions.

Additionally, carbon deposition occurs when catalysts treat exhaust gases with high carbon content. Carbon deposition deactivation refers to the physical phenomenon whereby hydrocarbons or other carbon sources generate carbonaceous deposits (e.g., coke) on catalyst surfaces or within pores during reactions, covering active sites, blocking pores, or encapsulating metal particles, thereby significantly reducing catalyst activity, selectivity, and stability.⁹¹ Jóźwiak et al.⁹² studied the activity and stability of single-metal Ni and Rh, and bimetallic Ni–Rh catalysts supported on silica for methane carbon dioxide reforming. Carbon deposition characteristics were investigated via H₂-temperature programmed reduction (H₂-TPR), temperature programmed oxidation (TPO), CO₂-TPR, FT-IR, etc. Results showed that the bimetallic 2.5Ni–2.5Rh/SiO₂ catalyst accumulated substantial carbon deposits (∼36%), contributing to reduced catalyst activity. Carbon deposition primarily relates to methane cracking (CH₄ → C + 2H₂) and carbon monoxide disproportionation (2CO → C + CO₂). These reactions generate substantial carbon. Carbon deposits physically cover noble metal nanoparticles, hinder reactant molecule adsorption, and significantly reduce active surface area, leading to substantial activity loss. Carbon deposition deactivation fundamentally involves shielding/blocking of active sites and mass transfer channels. Its essence is surface accumulation of reaction by-product carbon (particularly from cracking/disproportionation pathways). To enhance long-term stability under harsh conditions (e.g., high-carbon exhaust treatment), inhibiting/regulating these key carbon deposition pathways and reducing carbon generation/accumulation are equally crucial in optimizing reaction activity.

4. Strategies for inhibiting deactivation of Rh-based catalysts

Research on Rh catalyst deactivation mechanisms reveals that these deactivation pathways significantly impact catalyst efficiency in practical applications. Therefore, subsequent sections propose corresponding inhibition strategies for sulfur poisoning, NO_x poisoning, H₂O poisoning, and other chemical deactivation, as well as Rh sintering, support deactivation, and other physical deactivation. This provides theoretical guidance and technical support for developing high-performance Rh-based catalysts.

4.1 Strategies for inhibiting chemical deactivation

4.1.1 Sulfur poisoning inhibition strategy. In exhaust gas treatment, catalyst deactivation by sulfur dioxide primarily manifests as active site blockage and a decline in adsorption capacity, significantly impacting catalytic performance. To address this challenge, researchers have explored sulfur poisoning inhibition through various strategies: (1) catalyst design hinders sulfide adsorption. Liu et al.⁹³ designed a TiO₂–SiO₂ binary oxide support to evaluate catalyst sulfur resistance. Results showed that SiO₂ addition increased surface acid sites, weakened surface basicity, and inhibited sulfate formation, thereby enhancing sulfur resistance. (2) Acid or alkali catalyst treatment. Acid treatment primarily regenerates metal oxides, while alkali treatment regenerates non-metal oxides. Yang et al.⁹⁴ used H₂SO₄ to acid treat and regenerate a deactivated TiO₂–ZrO₂–CeO₂/ATS catalyst. Results indicated that acid treatment removes surface-deposited ammonium sulfate and metal oxides that damage acid centers and block channels, while increasing surface sulfate content and forming new acid centers. Treated catalysts exhibit good regeneration capacity. Yu et al.⁹⁵ proposed to regenerate the deactivated catalyst by alkali treatment with sodium hydroxide solution, and found that alkali treatment regeneration can effectively dissolve the deposited sulfate and improve the specific surface area of the catalyst, and the SCR activity of the regenerated catalyst can be restored to 74% of the initial activity. (3) Additive incorporation.⁹⁶ The first two methods increase development costs and production complexity. This prompted the development of simpler additive-based strategies, valued for operational simplicity and excellent sulfur resistance. As noted, Rh/Al₂O₃ catalysts treating sulfur-containing exhaust form aluminum sulfate and RhSO₄, reducing active sites. Researchers introduced Na promoters. XRD and ESCA revealed that added Na preferentially reacts with SO₂ to form Na₂SO₄. This competitive adsorption mechanism inhibits SO₂ adsorption on active sites, enhancing sulfur poisoning resistance.²² Almofleh et al.⁹⁷ used DFT to study boron doping effects on sulfur poisoning. Results showed that boron doping reduces sulfur binding energy on Rh surfaces, destabilizes sulfur bonds, and lowers rate-limiting step energy barriers. (4) Enhancing catalyst regeneration capacity represents another strategy to improve anti-deactivation capability. Feng et al.⁹⁸ prepared single-site catalyst RH₁/POPs by simply impregnating the dicarbonyl rhodium(I) precursor on a porous organic polymer. The catalyst is prone to sulfur poisoning in the presence of H₂S, resulting in a significant decline in catalytic activity. Hydrogen sulfide will combine with the active sites on the catalyst surface to form an inactive compound (SH)Rh(CO)(PPh3-frame)₂ thus inhibiting the reaction. Crucially, sulfur poisoning is reversible. After removing H₂S, the catalyst can restore its activity by itself. The inactive compound (SH)Rh(CO)(PPh3-frame)₂ in the catalyst can be reconverted into active HRh-(CO)(PPh3-frame)₂ under normal reaction conditions. This self-healing ability is mainly attributed to the metal–ligand interaction of the catalyst, which provides a reference for the design of poison resistant catalysts.

The above summarizes the current countermeasures researchers have taken to address sulfur poisoning in catalysts, and notes that certain achievements have been made. However, these strategies exhibit limitations. Although support modification and acid–base regeneration are effective, the former often involves complex synthesis processes, which increases the research and development and production costs; acid/alkali washing is a “remedial” measure that requires shutdown for regeneration, which is complicated and may produce secondary waste. In contrast, additives have become the most practical and widely adopted strategy due to relatively simple operation (typically introduced during preparation), significant sulfur resistance, and no disruption to continuous operation. However, existing promoter strategies require improvement regarding long-term catalyst performance/stability, and the universality of promoter effects necessitates further experimental verification and theoretical guidance. Future development will focus on creating more stable/efficient additives and designing catalyst structures with dynamic sulfur resistance or in situ regeneration capabilities, enabling a transition from “passive defense” to “active adaptation/self-healing” – an ideal solution for sulfur poisoning in complex environments.

4.1.2 NO_x poisoning inhibition strategy. NO_x is an important reactant in exhaust gas treatment, and catalyst poisoning caused by it requires attention. According to the NO_x poisoning mechanism (section 3.1.2), NO_x poisoning primarily results from strong NO_x adsorption on active sites and failure to timely remove oxygen (O) or nitrogen (N) species generated after NO_x dissociation. To address this challenge, researchers have developed various effective solutions: (1) support design. In the study of Hu et al.,⁹⁹ Rh was embedded in MOF-177 to improve the adsorption selectivity of NO at room temperature. Results showed that MOF-177 had limited NO saturation capacity (1.35 mL g⁻¹) and negligible selectivity, while 3 wt% Rh/MOF-177 increased the capacity to 16 mL g⁻¹. In simulated mixed gas, the NO adsorption capacity was 9–18 times higher than that of O₂ and CO₂. This highly selective support reduces active site occupation by other substances. (2) Bimetallic systems. Li et al.¹⁰⁰ studied the synergistic effect of Rh–Pt in three-way catalysis. Pt introduction increased the Rh surface concentration 2–3-fold versus the single-metal systems and effectively removed adsorbed oxygen species at low temperatures, facilitating O desorption from NO_x decomposition. The Rh–Pt catalyst's NO ignition temperature was 45 °C lower than that of the single-metal systems. (3) Rh microstructure regulation. Chen et al.¹⁰¹ prepared single-atom Rh, cluster, and nanoparticle catalysts via impregnation–reduction methods. Results showed that 1.9 nm Rh nanoparticles exhibited superior catalytic performance versus single-atom and cluster catalysts. In situ FTIR revealed that single-atom catalysts adsorbed NO and reacted slowly, hindering other gas adsorption and reaction efficiency. On Rh nanoparticles, NO preferentially adsorbs on Rh⁰ sites without hindering other gas adsorption, indicating that nanoparticle catalysts effectively prevent NO_x poisoning by strong adsorption, making them more active than single-atom catalysts.

The aforementioned NO_x poisoning suppression strategies are common and proven effective in most studies, but face challenges including: poor hydrothermal stability of MOF-177 supports, potentially limiting reactant transport; cost and complexity of bimetallic systems, where noble metal incorporation enhances performance but significantly increases material costs; additionally, more complex preparation processes for bimetallic catalysts (e.g., alloying degree and interface control) compared to single-metal systems; balance challenges in microstructure optimization, where the aforementioned optimal size may be condition-specific; significant challenges in achieving optimal size and maintaining stability during large-scale production, temperature cycling, vibration, and real operating conditions. Future development should focus on overcoming limitations and exploring efficient, stable, and low-cost routes such as: structural design optimization and multifunctional support development; exploring non-precious/low-precious metal combinations to reduce dependence while maintaining activity and anti-toxicity; developing advanced synthesis strategies (e.g., SMSI enhancement, surface coating/modification); precisely controlling Rh nanoparticle size (e.g., stabilizing at 1.5–2.5 nm) during preparation; imparting anti-sintering/anti-aggregation capabilities to ensure long-term structural stability.

4.1.3 H₂O poisoning inhibition strategy. To inhibit H₂O poisoning in Rh catalysts, the Rh/ZSM-5 catalyst prepared by Zhang et al.¹⁰² exhibited high H₂O resistance during reactions, with the high SiO₂/Al₂O₃ ratio (280) of ZSM-5 conferring hydrophobicity and weak acidity. These properties reduce strong H₂O adsorption on support surfaces, prevent excessive H₂O occupation of active sites, and thus mitigate H₂O poisoning effects. Sun et al.¹⁰³ constructed a superhydrophobic porous structure through a layered porous phosphite-based framework. After metallizing the superhydrophobic porous phosphite framework with Rh, the catalyst demonstrated better activity, significantly enhanced durability in simulated catalytic processes, and good recyclability compared to Rh catalysts on unmodified supports.

For inhibiting H₂O poisoning in Rh catalysts, researchers primarily adopted hydrophobic environment strategies, achieving some success in anti-H₂O-poisoning performance. However, existing hydrophobic strategies face challenges. Long-term chemical stability and mechanical durability of superhydrophobic structures under high-temperature oxidizing exhaust conditions remain uncertain; the pursuit of excessive hydrophobicity may impair reactant mass transfer/adsorption, while overemphasizing support hydrophobicity could weaken beneficial acidic properties or metal-anchoring capabilities. Future research should focus on developing high-temperature-stable robust hydrophobic materials, strengthening interface engineering/synergy, and achieving co-optimization of H₂O resistance, activity, and stability.

4.1.4 Other chemical deactivation inhibition strategies. As mentioned above, the catalyst will seriously affect the activity of the catalyst due to the change of the chemical state of Rh. For chemical-change-induced deactivation, researchers have proposed solutions including preparation process improvement, additive incorporation, and support optimization. (1) Preparation process improvement. Chen et al.¹⁰⁴ synthesized a Rh/CZ catalyst by a liquid-phase reduction adsorption method using glycerol as a reductant. Compared to traditional impregnation, this method maintained Rh in a reduced state: T₅₀ and T₉₀ for fresh/aged NO decreased by ∼20 °C, with post-aging Rh⁰ proportion at 32.3% versus 9.3% for traditional impregnation. (2) Additive incorporation. Kawabata et al.¹⁰⁵ studied the effect of La doping on a Rh/ZrO₂ catalyst. Results showed that La stabilizes Rh in a reduced state, enabling high catalytic activity despite weak oxygen storage capacity. Haneda et al.¹⁰⁶ systematically studied rare earth elements (La, Ce, Y, and Pr) on Rh/ZrO₂, finding that Y reduced NO conversion T₅₀ from 310 °C to 250 °C. Results indicated that Y₂O₃ strongly interacts with Rh, altering surface electronic states and promoting catalytic decomposition. It is conducive to the catalytic decomposition process. The optimization of the support can effectively inhibit the formation of compounds between the active components and support. (3) Support optimization. Li et al.¹⁰⁷ compared the performance differences of Rh/γ-Al₂O₃ and Rh/AlPO₄ catalysts. After thermal aging, Rh/γ-Al₂O₃ was severely deactivated due to inert Al₅Rh₂ alloy formation, while Rh/AlPO₄ showed better stability. Notably, aged Rh/AlPO₄ formed large Rh particles conducive to NO dissociation, exhibiting superior NO–CO catalytic activity. In addition to the above three common suppression strategies, Alikin et al.¹⁰⁸ provided us with another research idea. Aiming at the long-term stability of three-way catalysts, they focused on the regeneration mechanism of the Rh component after deactivation. The performance of Rh/Al₂O₃ and Rh/CeZrO₂ catalysts was tested by simulating engine braking (rapid cooling from 1200 °C to 600 °C air flow) and the standard aging mode. The results showed that the Rh/CeZrO₂ catalyst can be regenerated under high temperature oxidation conditions by simulating the rapid cooling of the engine during braking. This is because in the process of rapid cooling and regeneration, RhO species are further generated on the catalyst surface from Rh₂O₃, and the surface accessibility of rhodium species is significantly improved. The change of the oxidation state and the increase of dispersion are the main mechanisms of catalyst self-regeneration and activity recovery. Researchers have proposed various effective solutions for catalyst deactivation caused by Rh chemical state changes (e.g., oxidation and alloying). These strategies focus on stabilizing favorable Rh states (especially reduced Rh⁰) or inhibiting harmful compound formation. However, challenges persist. Improved processes like liquid-phase reduction adsorption show remarkable effects but involve cumbersome steps and harsh conditions, hindering scalability and transferability to other supports; long-term stability also requires verification. Rare earth additives (La and Y) stabilize the Rh valence and improve the electronic structure, but optimal dosage, support distribution, and long-term behavior (migration, enrichment, and pollutant interactions) need further study. Although AlPO₄ inhibits Al₅Rh₂, designing universally effective supports that prevent diverse harmful compounds (e.g., alloys and spinel phases) without sacrificing key properties remains difficult. AlPO₄'s applicability across all scenarios also requires consideration. Future development should therefore focus on efficient advanced preparation technologies and simpler, milder, and scalable synthetic methods. It should also involve rational additive/support design and control, deep mechanistic understanding of Rh stabilization and electronic control (e.g., charge transfer, isolation, and diffusion inhibition), and precise on-demand micro-addition to maximize effects and minimize side effects.

4.2 Physical deactivation inhibition strategy

4.2.1 Inhibition strategy of Rh sintering deactivation. In the process of exhaust gas treatment, the thermal deactivation resistance of Rh-based catalysts is the key factor to determine their long-term stable operation. Researchers have proposed a variety of solutions, such as the bimetallic system mentioned in section 4.1.2 for anti-sintering deactivation, adding a binder to inhibit the movement of Rh, and optimizing the support material and preparation process. These methods have been proved to be effective in inhibiting Rh sintering. (1) Bimetallic systems inhibit sintering. Vedyagin et al.¹⁰⁹ systematically studied the performance of a Pd–Rh/δ-Al₂O₃ bimetallic catalyst and found that its thermal stability was significantly better than that of the single metal systems. Specifically, the single-metal Rh catalyst is deactivated at high temperatures due to the diffusion of Rh³⁺ into the support, while the single-metal Pd catalyst is prone to particle agglomeration at 1000 °C. Through the construction of a Pd–Rh bimetallic system, the metal-to-metal interaction inhibits the aggregation of Pd and the diffusion of Rh³⁺ simultaneously. When the ratio of Pd : Rh was 3 : 2, the catalyst exhibited the best stability at 1000 °C. (2) Adding adhesive. It is mentioned in the introduction that the Rh/ZrO₂–CeO₂ catalyst will be encapsulated at high temperature. By adding γ-Al₂O₃ as the adhesive, the researchers achieved strong bonding between Rh and Al₂O₃, and the Rh particles were bound at or near the boundary, preventing irreversible complete encapsulation and avoiding catalyst deactivation even under complex and changing reaction conditions.¹¹⁰ (3) Optimize support materials. Jalal Samed et al.¹¹¹ developed a new and simple hydrothermal synthesis route for the preparation of ZrP₂O₇ from ZrO(OH)₂ and H₃PO₄ to use it as the active support of a Rh catalyst. The experimental results showed that Rh particles maintained a dispersed state of 5–10 nm after aging for 500 hours at 900 °C in air with 10% H₂O and retained a stable structure at 1200 °C. (4) Improve the preparation process. Misumi et al.¹¹² used pulsed arc plasma and RF magnetron sputtering technology to prepare an Rh coating with a surface thickness of 7 nm, and its catalytic performance was 50 times higher than that of the traditional slurry coating. The catalyst prepared by RF magnetron sputtering technology was significantly deactivated due to poor dispersion of Rh (low coverage) and oxidation of Rh. The Rh coating prepared by pulsed arc plasma has excellent thermal stability and sintering resistance due to its strong adhesion to the metal surface. The four inhibition strategies are all implemented during the material preparation process. Alikin et al.¹¹³ proposed a regeneration strategy for catalysts after they have aged. They studied the regeneration ability of a Rh/ZrCeYLaO₂ catalyst after high temperature treatment. The research results indicated that the catalytic activity of the Rh/ZrCeYLaO₂ catalyst, which has undergone high-temperature aging, can be significantly restored through rapid cooling after high-temperature oxidation (simulating fuel cut-off). Following rapid cooling, some of the agglomerated Rh particles redisperse onto the oxide surface, enhancing both the quantity and distribution of active sites. Moreover, this regenerated state maintains excellent stability and repeatability even after repeated high-temperature hydrothermal treatments. Improving the thermal deactivation resistance (primarily sintering and encapsulation) of Rh-based catalysts is critical. Researchers have developed various effective strategies. However, limitations remain: bimetallic systems are costly; binders may impede mass transfer or induce side reactions; the universality and large-scale preparation feasibility of novel supports (e.g., ZrP₂O₇) require verification; advanced technologies (e.g., plasma) involve complex equipment and high costs, and sputtered layers are prone to oxidative deactivation. Future efforts should focus on: developing low-cost, high-stability alternative materials; exploring non-precious metal additives or novel high-temperature-resistant, high-surface-area supports to reduce costs while suppressing sintering; enhancing interface engineering and structural design to improve thermal stability through precise modulation of strong metal–support interactions (SMSI) or by designing core–shell and anchoring site structures; optimizing and advancing efficient preparation technologies – improving processes such as pulsed arc plasma to enhance efficiency, reduce energy consumption and costs, and broaden applicability; exploring more convenient, large-scale methods to prepare catalytic layers with strong adhesion and oxidation resistance. In summary, although existing strategies effectively inhibit thermal deactivation, cost, process complexity, and long-term stability remain bottlenecks. Future breakthroughs require integrating low-cost/high-stability material development, precise interface control, and efficient/large-scale preparation technologies to achieve long-term stable operation of Rh-based catalysts under harsh conditions.

4.2.2 Support deactivation inhibition strategy. In practical production processes, besides Rh sintering, researchers have proposed solutions addressing performance deterioration caused by the thermodynamic behavior of supports at high temperatures. For instance, pore structure changes induced by support sintering can be mitigated through novel technologies enabling catalyst regeneration. Support phase transitions can be suppressed by adding additives and optimizing preparation processes. (1) Adopt new regeneration technology. Lin et al.¹¹⁴ regenerated spent catalysts using microwave heating combined with ultrasonic spray pyrolysis. Experimental data indicated that the specific surface area of spent catalysts increased from 293 m² g⁻¹ to 1263 m² g⁻¹, with a total pore volume of 1.28 mL g⁻¹. Ultrasonic spray pyrolysis promotes abundant micropore and mesopore formation. (2) Add additives. Rossignol et al.¹¹⁵ systematically investigated the effect of sol–gel-introduced dopants on alumina thermal stability. Results demonstrated that 1–2 mol% Ba or Pr doping increased the θ-to-α phase transition temperature to 1315 °C, delaying phase transformation. Alikin et al.¹¹⁶ added BaO to a variety of catalysts (single-metal and multi-metal) supported on ZrO₂ doped Al₂O₃. XRD data indicated that adding 3 wt% BaO can significantly suppress the formation of the α-Al₂O₃ phase in a single-metal Rh-loaded catalyst at high temperatures (1100 °C). This is likely due to the synergistic interaction between BaO and the surface or lattice of alumina, which generates a steric hindrance effect. This effect impedes the sintering and reorganization of alumina grains, thereby delaying or inhibiting the high-temperature phase transition of α-Al₂O₃. This can maintain a high specific surface area and pore volume, thereby improving the thermal stability and catalytic performance of the catalyst. Stoyanovskii et al.^117,118 studied the effect of La doping on a Rh/Al₂O₃ catalyst. By electron paramagnetic resonance (EPR) spin probing, X-ray diffraction (XRD), photoluminescence (PL) and other techniques, the surface and bulk structure of the catalyst, as well as the migration and localization of Rh, were systematically studied. The experimental results showed that La doping can improve the thermal stability and surface area of the catalyst at low loading (0.08 wt%). From 800 °C, a LaAl₁₁O₁₈ octahedral structure will be formed, which contributes to the thermal stability of aluminum oxide. At the same time, to some extent, it reduces the formation of Rh and Al₂O₃ into RhAlOx, which inhibits the catalyst deactivation due to the formation of RhAlOx. Moreover, La doping can stabilize the γ-Al₂O₃ structure at high temperature (800–1000 °C), prevent its transformation to α-Al₂O₃, and affect the spatial diffusion and localization of Rh³⁺ ions, effectively preventing Rh from being irreversibly wrapped and inactivated. However, when the Rh loading is high (0.4 wt%), La doping will accelerate the diffusion process of Rh³⁺ ions to the bulk phase of alumina, and make Rh close to La ions in a LaAl₁₁O₁₈ structure, resulting in the residual amount of Rh on the surface equivalent to that of the alumina without La doping, but the catalytic activity decreases faster, resulting in faster deactivation of the catalyst and lower overall stability. Therefore, when doping rare earth elements, in addition to the positive impact on the stability or activity of the catalyst, we should also consider the possible negative impact on precious metals. (3) Optimize the preparation process. Jang et al.¹¹⁹ developed an undoped alumina preparation strategy. In this method, a rod-shaped boehmite precursor was synthesized by a hydrothermal method and then processed via centrifugation (Al₂O₃-C), oven drying (Al₂O₃-O), or freeze drying (Al₂O₃-F) to control alumina's primary exposed crystal planes and particle size. After treatment at 1200 °C, Al₂O₃-F retained 80% specific surface area, initiated α-Al₂O₃ phase transition at 1200 °C, and required 1300 °C for completion – significantly higher than conventional materials.

To address support performance degradation caused by sintering, pore collapse, and phase transitions at high temperatures, researchers have proposed innovative regeneration technologies, additive incorporation to suppress phase transitions, and precursor treatment optimization for structural control. However, existing methods require further improvement: advanced regeneration technologies (e.g., microwave-ultrasound) involve complex equipment, high costs, and limited scalability; dopants may increase costs, and their potential loss or inefficacy under prolonged high-temperature exposure requires verification; freeze-drying and similar specialized processes exhibit high energy consumption, low efficiency, and scalability challenges; most studies focus on individual supports (e.g., Al₂O₃) and lack universal solutions for others (e.g., ZrO₂–CeO₂). Future research should develop low-cost, efficient, and scalable regeneration/stabilization technologies and design novel supports with intrinsic sintering/phase-transition resistance requiring no additional treatment. Support design must simultaneously consider thermal stability, pore-structure preservation, active-component compatibility, and deactivation resistance. In summary, effective strategies exist for high-temperature degradation; however, cost, scalability, and universality remain bottlenecks. Future breakthroughs require economical/efficient technology development, intrinsically high-stability material design, and scalable process optimization to achieve long-term catalyst support stability under harsh conditions.

4.2.3 Other physical deactivation inhibition strategies. Regarding catalyst deactivation induced by thermal stress, instances caused solely by thermal stress are less common and are predominantly accompanied by thermal sintering and material segregation. Methods to mitigate thermal stress reported in most studies include optimizing calcination processes,¹²⁰ component control,¹²¹ support control,¹²²etc., which usually avoids the influence of thermal stress while solving the problem of improving the thermal stability of thermal sintering.

To mitigate catalyst carbon deposition, additives can be incorporated, such as oxides (e.g., MgO and CaO) that enhance support surface basicity, promote CO₂ adsorption, generate reactive oxygen species, and retard surface carbon deposition.¹²³ Alternatively, adding rare-earth oxides (e.g., CeO₂ and La₂O₃) with abundant oxygen vacancies enables frequent oxygen storage/release to catalyze carbon oxidation.¹²⁴ This can also be designed through the support. Du et al.¹²⁵ reported a CRM catalyst of HT-NiMgAl with a multistage pore structure. This hierarchical pore configuration enhances specific surface area and improves particle dispersion. Due to its restriction on the active component region, it can effectively inhibit carbon deposition. Process optimization further inhibits carbon deposition, including adjusting gas ratios (e.g., higher CO₂/CH₄) to thermodynamically favor carbon oxidation and introducing O₂ (direct carbon oxidation) or H₂O (carbon gasification via hydroxyl radical generation).¹²⁶

Strategies to mitigate catalyst deactivation primarily include additive incorporation, support design, and process optimization. These approaches enhance catalyst resistance to carbon deposition by improving oxidative carbon removal capacity or physically suppressing carbon formation/accumulation. However, challenges persist: basic additives may conflict with other components (e.g., acid sites) or degrade at high temperatures; rare-earth oxides may exhibit diminished oxygen storage/release capacity under prolonged high-temperature exposure; hierarchical porous supports face preparation complexity and mechanical strength limitations; process optimization (e.g., increasing CO₂ concentration or adding O₂/H₂O) may elevate energy consumption or alter reaction pathways, compromising efficiency and cost-effectiveness. Future work should focus on developing multifunctional carbon-resistant supports integrated with in situ monitoring and process control technologies. This integration aims to enable efficient, low-consumption in situ regeneration while suppressing carbon deposition, ultimately overcoming the trade-offs among long-term efficacy, cost, and side effects inherent in current strategies.

5. Limitations and prospects of Rh-based catalysts

5.1 Limitations in the development of Rh-based catalysts

Although Rh-based catalysts show excellent catalytic performance in the field of exhaust gas purification, the deactivation problem has always been the key bottleneck restricting their industrial application. Rh catalyst deactivation mainly includes sulfur poisoning, NO_x poisoning, H₂O poisoning and other chemical deactivation, and Rh sintering, support deactivation and other physical deactivation. Although the existing research can partially alleviate the deactivation problem through the introduction of additives and the construction of bimetallic systems, these methods generally have limitations such as the complexity of the system and the rising cost of preparation and are difficult to fundamentally solve the problem of catalyst deactivation.

Most of the existing studies focus on single catalyst systems and lack systematic comparative studies on different support materials (such as Al₂O₃ and CeO₂), active components and preparation methods. Secondly, most experiments were conducted under simplified conditions, and the reaction conditions were too simple, which was different from the complex situation of the multipollutant coexistence and working condition fluctuation in the actual industrial environment, which limited the actual reference value of the research results. Although some Rh-based catalysts exhibit reversible deactivation, regeneration under synergistic deactivation mechanisms remains insufficiently studied. In particular, the key issues such as the degree of activity recovery under complex deactivation conditions, the optimization of regeneration processes and long-term stability still need to be further explored. In addition, the contradiction among the preparation process, performance and cost of Rh catalysts is also a difficult problem to solve, such as: high-performance Rh catalysts depend on the size of nanoparticles and uniform dispersion. Although the traditional impregnation method can achieve low-cost mass production, it is difficult to accurately control the particle size and dispersion, which will lead to a dead cycle in which the pursuit of high dispersion requires complex processes, resulting in rising costs, while simplifying the process will reduce the activity. As Rh is a rare and precious metal, its high price limits its promotion in large-scale industrial applications. The preparation cost can be reduced by introducing bimetallic composite systems (such as Pt and Pd, the price is lower than Rh) or reducing the Rh content of catalysts, but these methods often sacrifice the catalytic performance. Therefore, it is still of great significance to develop catalysts with better performance, stronger deactivation resistance and lower cost.

5.2 Prospects

In recent years, the research of catalyst support materials has expanded from traditional metal inorganic materials to new functional material systems. The research shows that graphene, metal organic frameworks (MOFs) and other new catalyst support materials show significant advantages in specific surface area, structural adjustability, and selective adsorption with their unique physical and chemical properties, which provides a new way to improve the comprehensive performance of catalysts. Taking MOF-177 as an example, the material has ultra-low density, large specific surface area and a regular pore structure, showing excellent thermal stability and gas adsorption selectivity. It is worth noting that MOF-177 has weak adsorption of O₂ and N₂, and can be used as a good NO adsorption support.¹²⁷ Yang et al.¹²⁸ constructed a new Rh/MOF-177 catalyst by limiting 3 wt% Rh nanoparticles in the MOF-177 channel. The experimental results showed that the NO adsorption capacity of the catalyst was 9–18 times higher than that of the traditional support, and the selectivity coefficients for NO/CO₂ and NO/O₂ were increased by about 18 times. This selective adsorption property effectively inhibits the occupation of active sites by competitive molecules such as CO₂ and O₂ and significantly reduces the poisoning probability of the catalyst. XPS characterization revealed that the strong interaction between Rh nanoparticles and MOF-177 coordination unsaturated metal sites was the key factor to improve the stability of the catalyst.

The development of computational materials science also provides a new research direction for catalyst research. Through molecular simulation, first principles calculation (such as density functional theory (DFT)) and machine learning, computational materials science can efficiently and deeply reveal the catalytic reaction mechanism (such as the active site structure, reaction path and energy barrier), predict the intrinsic performance of catalyst materials (such as adsorption energy, electronic structure and stability), and guide the design and screening of new high-performance catalysts, so as to significantly accelerate the research and development process, reduce the cost of experimental trial and error, and deepen the understanding of the catalytic mechanism. Tan et al.¹²⁹ systematically studied the catalytic mechanism of Rh (100) and Rh (111) crystal surfaces for the NO–CO reaction using density functional theory (DFT). The calculation results show that (100) has a better ability to stabilize all related reaction transition states and adsorb reactants. Liu et al.¹³⁰ further constructed a micro kinetic model and quantitatively analyzed the catalytic performance difference between the two crystal planes. The simulation results show that under the same conditions, the catalytic activity and product selectivity of the Rh (100) surface are better than those of the Rh (111) surface, and the activation energy of (100) is relatively low, which makes the reaction easier. This research method combining theoretical calculation with experimental verification not only reveals the influence of the mechanism of the crystal plane effect but also provides an important scientific basis for the design of efficient Rh-based catalysts.

In addition, according to the research, metal–support strong interaction (SMSI) is an important concept in heterogeneous catalysis, which can significantly improve the stability of the catalyst and may change the performance of the catalyst. However, the SMSI effect is generally reversible and may fade under specific redox conditions, which limits its application environment.¹³¹ Therefore, the performance of the catalyst can be optimized by constructing/adjusting the SMSI effect. Li et al.¹³² constructed a new SMSI system under mild conditions by a mechanochemical method, which broke through the traditional preparation conditions requiring high temperature and oxidation/reduction atmospheres, and could be prepared in a shorter time and on more supports, and the degree of encapsulation can be controlled by the reducibility of the reductant. The activity and stability of the catalyst were significantly improved.

6. Conclusion

Rh-based catalysts have become key materials in this field due to their excellent exhaust treatment performance. However, their practical application is limited by complex deactivation mechanisms. This paper systematically reviews deactivation mechanisms and inhibition strategies, as shown in Fig. 17. Key conclusions are:

Fig. 17 An overview of this review. Left: Catalyst deactivation, including chemical deactivation and physical deactivation, chemical deactivation mainly refers to “chemical structural changes” and “poisoning”; physical deactivation mainly involves “sintering” and “phase transitions”. Right: Deactivation inhibition strategies: including bimetallic system, structural modification, additive incorporation, and preparation process optimization.

(1) The main reactions of Rh-based catalysts in exhaust treatment include SCR and DND reactions. Among these, SCR is the most widely used technology, categorizable as NH₃-SCR, H₂-SCR, CO-SCR, etc. based on the reductant type. Although currently it is the most effective exhaust treatment method, it still faces challenges including high system complexity and cost. The DND reaction is more environmentally favorable. It proceeds via direct NO decomposition without additional reagents. However, it suffers from low catalytic efficiency and scalability challenges.

(2) The main reaction mechanisms of Rh-based catalysts during exhaust gas catalytic purification include Eley–Rideal (E–R), Langmuir–Hinshelwood (L–H), and Mars–van-Krevelen (MvK) mechanisms. Among these, the MvK mechanism is widely considered dominant for exhaust gas catalytic degradation over Rh-based catalysts. These mechanisms may coexist in different reaction systems with synergistic or competitive relationships: L–H or E–R mechanisms typically dominate catalyst surface reactions at lower temperatures, while the MvK mechanism predominates at higher temperatures and involves lattice oxygen participation.

(3) The chemical deactivation of Rh-based catalysts primarily arises from sulfur poisoning, NO_x poisoning, H₂O poisoning, and other mechanisms. Sulfur, NO_x, and H₂O poisoning predominantly results from active metal site blockage by substances and reactants in the reaction atmosphere through strong chemical adsorption, leading to catalyst deactivation. Other chemical deactivation mainly stems from changes in the Rh chemical state (metallic Rh to oxidized Rh) and reactions with supports (e.g., Rh with Al₂O₃ forming RhAlO_x), which reduce surface-active sites and cause catalyst deactivation.

(4) The physical deactivation of Rh-based catalysts primarily results from Rh sintering, support deactivation, and other physical deactivation mechanisms. Rh sintering mainly involves high-temperature growth and migration of Rh particles, causing substantial active site reduction, while support deactivation encompasses support sintering and phase transformation. Both phenomena stem from high-temperature structural changes and reduced specific surface area of support materials, leading to a significant decline in catalyst performance. Other physical deactivation mechanisms include thermal stress and carbon deposition, which also cause catalyst structural damage and active site coverage.

(5) For chemical and physical catalyst deactivation, researchers have proposed numerous improvement strategies, including bimetallic system construction, structural modification, additive incorporation, and preparation process optimization. Deactivation was mitigated and stability was enhanced to some extent.

Current research primarily focuses on catalyst deactivation mechanisms and higher-performance catalyst development, but Rh-based catalyst regeneration studies remain relatively scarce. Catalyst regeneration is crucial for maximizing industrial catalyst utilization and cost reduction. Therefore, more comprehensive research is needed to develop green, efficient, simple-to-operate, and low-cost catalyst regeneration technologies.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Acknowledgements

This research was financially supported by the Scientific and Technological Project of Yunnan Precious Metals Laboratory (YPML-2023050232 & YPML-20240502055) and the Science Research Project of Yunnan Province (202502AB080011 & 202303AP140006). The authors would like to thank the Instrumental Analysis Center of KUST for sample characterization.

References

1. X. Chen, H. Zhao, J. Tang, M. Cui and X. Qiao, Deep catalytic oxidation of chlorotoluene in waste gas with Ce-Mn bimetallic oxide, Huagong Huanbao, 2010, 30, 465–468.
2. X. Sun, R. T. Guo, M. Y. Li, P. Sun, W. G. Pan, S. M. Liu, J. Liu and S. W. Liu, The promotion effect of Fe on CeZr₂O_x catalyst for the low-temperature SCR of NO_x by NH₃, Res. Chem. Intermed., 2018, 44(5), 3455–3474.
3. Y. Yokoi and H. Uchida, Catalytic activity of perovskite-type oxide catalysts for direct decomposition of NO: Correlation between cluster model calculations and temperature-programmed desorption experiments, Catal. Today, 1998, 42(1), 167–174.
4. J. Liu, Z. Zhao, C. M. Xu and A. J. Duan, Simultaneous removal of NO_x and diesel soot over nanometer Ln-Na-Cu-O perovskite-like complex oxide catalysts, Appl. Catal., B, 2008, 78(1), 61–72.
5. Q. Yuan, Z. Zhang, N. Yu and B. Dong, Cu-ZSM-5 zeolite supported on SiC monolith with enhanced catalytic activity for NH₃-SCR, Catal. Commun., 2018, 108, 23–26.
6. H. Xu, L. Zhu, Q. Wu, X. Meng and F. S. Xiao, Advances in the synthesis and application of the SSZ-39 zeolite, Inorg. Chem. Front., 2022, 9(6), 1047–1057.
7. Z. R. Ismagilov, E. V. Matus and L. T. Tsikoza, Direct conversion of methane on Mo/ZSM-5 catalysts to produce benzene and hydrogen: achievements and perspectives, Energy Environ. Sci., 2008, 1(5), 526–541.
8. J. Zhu, Y. Wei, W. Chen, Z. Zhao and A. Thomas, Graphitic carbon nitride as a metal-free catalyst for NO decomposition, Chem. Commun., 2010, 46(37), 6965–6967.
9. J. Zawadzki and M. Wiśniewski, Adsorption and decomposition of NO on carbon and carbon-supported catalysts, Carbon, 2002, 40(1), 119–124.
10. J. Yang, G. Mestl, D. Herein, R. Schlögl and J. Find, Reaction of NO with carbonaceous materials: 1. Reaction and adsorption of NO on ashless carbon black, Carbon, 2000, 38(5), 715–727.
11. S. Xu, Y. Zhang, C. Hardacre and Z. Liu, Enhanced catalytic activity of Pd supported on TiO₂ nanowire for the H₂-SCR of NO_x in the presence of oxygen, ACS Sustainable Chem. Eng., 2023, 11(28), 10453–10461.
12. I. A. Resitoglu, A. Keskin, H. Özarslan and H. Bulut, Selective catalytic reduction of NO_x emissions by hydrocarbons over Ag-Pt/Al₂O₃ catalyst in diesel engine, Int. J. Environ. Sci. Technol., 2019, 16(11), 6959–6966.
13. H. Li, P. Xiao, T. Wang, J. Zhu and J. Li, Recent progress on catalysts used for NO decomposition, Zhongguo Kexue: Huaxue, 2014, 44(12), 1951–1965.
14. J. H. Wang, H. Chen, Z. C. Hu, M. F. Yao and Y. D. Li, A review on the Pd-based three-way catalyst, Catal. Rev.: Sci. Eng., 2015, 57(1), 79–144.
15. S. Nandi, P. Arango, C. Chaillou, C. Dujardin, P. Granger, E. Laigle, A. Nicolle, C. Norsic and M. Richard, Relationship between design strategies of commercial three-way monolithic catalysts and their performances in realistic conditions, Catal. Today, 2022, 384, 122–132.
16. C. H. Bartholomew, Mechanisms of catalyst deactivation, Appl. Catal., A, 2001, 212(1–2), 17–60.
17. N. M. Martin, P. Velin, M. Skoglundh, M. Bauer and P. A. Carlsson, Catalytic hydrogenation of CO₂ to methane over supported Pd, Rh and Ni catalysts, Catal. Sci. Technol., 2017, 7(5), 1086–1094.
18. P. Granger and V. I. Parvulescu, Catalytic NO_x abatement systems for mobile sources: From three-way to lean burn after-treatment technologies, Chem. Rev., 2011, 111(5), 3155–3207.
19. S. Q. Sun, C. X. Jin, W. Z. He, G. M. Li, H. C. Zhu and J. W. Huang, A review on management of waste three-way catalysts and strategies for recovery of platinum group metals from them, J. Environ. Manage., 2022, 305, 11438.
20. M. Shelef and G. W. Graham, Why rhodium in automotive three-way catalysts?, Catal. Rev.: Sci. Eng., 1994, 36(3), 433–457.
21. Y. Li, A. Sundermann, O. Gerlach, K. B. Low, C. C. Zhang, X. Zheng, H. Zhu and S. Axnanda, Catalytic decomposition of N₂O on supported Rh catalysts, Catal. Today, 2020, 355, 608–619.
22. F. Y. Chang, J. C. Chen and M. Y. Wey, Activity and characterization of Rh/Al₂O₃ and Rh-Na/Al₂O₃ catalysts for the SCR of NO with CO in the presence of SO₂ and HCl, Fuel, 2010, 89(8), 1919–1927.
23. K. A. Almusaiteer and S. S. C. Chuang, Infrared characterization of Rh surface states and their adsorbates during the NO-CO reaction, J. Phys. Chem. B, 2000, 104(10), 2265–2272.
24. M. Machida, Y. Uchida, Y. Ishikawa, S. Hinokuma, H. Yoshida, J. Ohyama, Y. Nagao, Y. Endo, K. Iwashina and Y. Nakahara, Thermostable Rh metal nanoparticles formed on Al₂O₃ by high-temperature H₂ reduction and its impact on three-way catalysis, J. Phys. Chem. C, 2019, 123(40), 24584–24591.
25. Y. Zhang, P. Glarborg, M. P. Andersson, K. Johansen, T. K. Torp, A. D. Jensen and J. M. Christensen, Sulfur poisoning and regeneration of Rh-ZSM-5 catalysts for total oxidation of methane, Appl. Catal., B, 2020, 277, 119176.
26. K. J. Kim, Y. L. Lee, G. R. Hong, S. Y. Ahn, B. J. Kim, S. S. Lee, Y. Jeon and H.-S. Roh, A study on the activity recovery behavior of noble metal catalysts against sulfur poisoning, Catal. Today, 2024, 425, 114361.
27. V. I. Pârvulescu, P. Grange and B. Delmon, Catalytic removal of NO, Catal. Today, 1998, 46(4), 233–316.
28. M. Ikeda, T. Tago, M. Kishida and K. Wakabayashi, Thermal stability of an SiO₂-coated Rh catalyst and catalytic activity in NO reduction by CO, Chem. Commun., 2001, 2512–2513.
29. S. Bernal, F. J. Botana, J. J. Calvino, G. A. Cifredo, J. A. Pérez-Omil and J. M. Pintado, HREM study of the behaviour of a Rh/CeO₂ catalyst under high temperature reducing and oxidizing conditions, Catal. Today, 1995, 23(3), 219–250.
30. D. M. Fernandes, C. F. Scofield, A. A. Neto, M. J. B. Cardoso and F. M. Z. Zotin, Thermal deactivation of Pt/Rh commercial automotive catalysts, Chem. Eng. J., 2010, 160(1), 85–92.
31. Y. Hu, K. Griffiths and P. R. Norton, Surface science studies of selective catalytic reduction of NO: Progress in the last ten years, Surf. Sci., 2009, 603(10), 1740–1750.
32. S. Jones, Y. Ji, A. Bueno-Lopez, Y. Song and M. Crocker, CeO₂-M₂O₃ passive NO_x adsorbers for cold start applications, Emiss. Control Sci. Technol., 2017, 3(1), 59–72.
33. C. X. Liu, H. J. Wang, Z. Y. Zhang and Q. L. Liu, The latest research progress of NH₃-SCR in the SO₂ resistance of the catalyst in low temperatures for selective catalytic reduction of NO_x, Catalysts, 2020, 10(9), 1034.
34. Z. W. Shi, Q. G. Peng, J. Q. E., J. Wei, R. Yin and R. X. Fu, Mechanism, performance and modification methods for NH₃-SCR catalysts: A review, Fuel, 2023, 331, 125885.
35. Y. Guan, Y. Liu, Q. Lv, B. Wang and D. Che, Review on the selective catalytic reduction of NO_x with H₂ by using novel catalysts, J. Environ. Chem. Eng., 2021, 9(6), 106770.
36. Z. M. Liu, J. H. Li and S. I. Woo, Recent advances in the selective catalytic reduction of NO_x by hydrogen in the presence of oxygen, Energy Environ. Sci., 2012, 5(10), 8799–8814.
37. R. Burch, A. A. Shestov and J. A. Sullivan, A steady-state isotopic transient kinetic analysis of the NO/O₂/H₂ reaction over Pt/SiO₂ catalysts, J. Catal., 1999, 188(1), 69–82.
38. Z. C. Xu, Y. R. Li, Y. T. Lin and T. Y. Zhu, A review of the catalysts used in the reduction of NO by CO for gas purification, Environ. Sci. Pollut. Res., 2020, 27(7), 6723–6748.
39. W. L. Huang, S. Y. Yang, X. D. Li, C. L. Ding, S. Ren, S. C. Li, H. R. Zhang, X. Y. Pi and X. T. Yin, Performance, reaction mechanism and modification methods for Mn-based CO-SCR catalysts: A review, J. Environ. Chem. Eng., 2024, 12(5), 113593.
40. P. P. Xie, W. X. Ji, Y. D. Li and C. J. Zhang, NO direct decomposition: Progress, challenges and opportunities, Catal. Sci. Technol., 2021, 11(2), 374–391.
41. Y. Teraoka, K. Torigoshi, H. Yamaguchi, T. Ikeda and S. Kagawa, Direct decomposition of nitric oxide over stannate pyrochlore oxides: Relationship between solid-state chemistry and catalytic activity, J. Mol. Catal. A: Chem., 2000, 155(1–2), 73–80.
42. S. Fang, A. Takagaki, M. Watanabe and T. Ishihara, The direct decomposition of NO into N₂ and O₂ over copper doped Ba₃Y₄O₉, Catal. Sci. Technol., 2020, 10(8), 2513–2522.
43. T. Ishihara, M. Ando, K. Sada, K. Takiishi, K. Yamada, H. Nishiguchi and Y. Takita, Direct decomposition of NO into N₂ and O₂ over La(Ba)Mn(In)O₃ perovskite oxide, J. Catal., 2003, 220(1), 104–114.
44. P. Stelmachowski, F. Zasada, W. Piskorz, A. Kotarba, J. F. Paul and Z. Sojka, Experimental and DFT studies of N₂O decomposition over bare and Co-doped magnesium oxide-insights into the role of active sites topology in dry and wet conditions, Catal. Today, 2008, 137(2–4), 423–428.
45. P. Pietrzyk, F. Zasada, W. Piskorz, A. Kotarba and Z. Sojka, Computational spectroscopy and DFT investigations into nitrogen and oxygen bond breaking and bond making processes in model deNO_x and deN₂O reactions, Catal. Today, 2007, 119(1), 219–227.
46. S. Tsujimoto, K. Yasuda, T. Masui and N. Imanaka, Effects of Tb and Ba introduction on the reaction mechanism of direct NO decomposition over C-type cubic rare earth oxides based on Y₂O₃, Catal. Sci. Technol., 2013, 3(8), 1928–1936.
47. S. Yang, Y. Fu, Y. Liao, S. Xiong, Z. Qu, N. Yan and J. Li, Competition of selective. catalytic reduction and non selective catalytic reduction over MnO_x/TiO₂ for NO removal: the relationship between gaseous NO concentration and N₂O selectivity, Catal. Sci. Technol., 2014, 4(1), 224–232.
48. D. H. Kim, Sulfation and desulfation mechanisms on Pt-BaO/Al₂O₃ NO_x storage-reduction (NSR) catalysts, Catal. Surv. Asia, 2014, 18(1), 13–23.
49. Q. Zheng, R. Farrauto, M. Deeba and I. Valsamakis, Part I: A comparative thermal aging study on the regenerability of Rh/Al₂O₃ and Rh/Ce_xO_y-ZrO₂ as model catalysts for automotive three way catalysts, Catalysts, 2015, 5(4), 1770–1796.
50. F. Y. Gao, X. L. Tang, H. H. Yi, S. Z. Zhao, C. L. Li, J. Y. Li, Y. R. Shi and X. M. Meng, A review on selective catalytic reduction of NO_x by NH₃ over Mn-based catalysts at low temperatures: Catalysts, mechanisms, kinetics and DFT calculations, Catalysts, 2017, 7(7), 199.
51. X. Li, S. Ren, Y. Jiang, Z. Chen, L. Wang, M. Liu and T. Chen, In situ IR spectroscopy study of NO removal over CuCe catalyst for CO-SCR reaction at different temperature, Catal. Today, 2023, 418, 114082.
52. C. Cai, J. Chen, W. Fu, S. Ning, H. Zhou, M. Kashif and Y. Su, Study on CO-SCR performance of Ce modified OMS-2 catalysts, J. Environ. Chem. Eng., 2023, 11(6), 111589.
53. J. He and C. Yin, Research progress of catalysts for selective catalytic reduction of nitrogen oxides by hydrogen, Yingyong Huagong, 2020, 49(11), 2906.
54. Y. Pan, N. Li, K. Li, S. Ran, C. Wu, Q. Zhou, J. Liu and S. Li, Enhanced low-temperature behavior of selective catalytic reduction of NO_x by CO on Fe-based catalyst with looping oxygen vacancy, Chem. Eng. J., 2023, 461, 141814.
55. A. J. Martín, S. Mitchell, C. Mondelli, S. Jaydev and J. Pérez-Ramírez, Unifying views on catalyst deactivation, Nat. Catal., 2022, 5(10), 854–866.
56. M. S. Wilburn and W. S. Epling, Formation and decomposition of sulfite and sulfate species on Pt/Pd catalysts: An SO₂ oxidation and sulfur exposure study, ACS Catal., 2019, 9(1), 640–648.
57. M. A. Ocsachoque, J. I. E. Russman, B. Irigoyen, D. Gazzoli and M. G. González, Experimental and theoretical study about sulfur deactivation of Ni/CeO₂ and Rh/CeO₂ catalysts, Mater. Chem. Phys., 2016, 172, 69–76.
58. C. Wang, X. Hou, L. Jin, J. Li, L. Gu and L. Yang, Review on the impact of SO₂ on VOCs oxidation: Mechanisms and anti-poisoning strategies, Fuel, 2024, 359, 130450.
59. K. I. Hadjiivanov, Identification of neutral and charged N_xO_y surface species by IR spectroscopy, Catal. Rev.: Sci. Eng., 2000, 42(1–2), 71–144.
60. A. M. de Oliveira, I. M. Baibich, N. R. C. F. Machado, M. L. Mignoni and S. B. C. Pergher, Decomposition of nitric oxide on Pd-mordenite, Catal. Today, 2008, 133–135, 560–564.
61. Y. X. Wang, R. Oord, D. van den Berg, B. M. Weckhuysen and M. Makkee, Oxygen vacancies in reduced Rh/ and Pt/Ceria for highly selective and reactive reduction of NO into N₂ in excess of O₂, ChemCatChem, 2017, 9(15), 2935–2938.
62. Y. Li, M. Q. Shen, J. Q. Wang, T. M. Wan and J. Wang, Influence of sulfation and regeneration on Pt/Al₂O₃ for NO oxidation, Catal. Sci. Technol., 2015, 5(3), 1731–1740.
63. W. B. Bae, D. Y. Kim, S. W. Byun, S. J. Lee, S. K. Kuk, H. J. Kwon, H. C. Lee, M. J. Hazlett, C. Liu, Y. J. Kim, M. Kim and S. B. Kang, Direct NO decomposition over Rh-supported catalysts for exhaust emission control, Chem. Eng. J., 2023, 475, 146005.
64. H. Zhu, Y. Li and X. Zheng, In situ DRIFTS study of CeO₂ supported Rh catalysts for N₂O decomposition, Appl. Catal., A, 2019, 571, 89–95.
65. H. Yang, C. Ma, X. Zhang, Y. Li, J. Cheng and Z. Hao, Understanding the active sites of Ag/Zeolites and deactivation mechanism of ethylene catalytic oxidation at Room Temperature, ACS Catal., 2018, 8(2), 1248–1258.
66. Q. Dai, L. L. Yin, S. Bai, W. Wang, X. Wang, X. Q. Gong and G. Lu, Catalytic total oxidation of 1,2-dichloroethane over VO_x/CeO₂ catalysts: Further insights via isotopic tracer techniques, Appl. Catal., B, 2016, 182, 598–610.
67. A. Z. Abdullah, M. Z. A. Bakar and S. Bhatia, Combustion of chlorinated volatile organic compounds (VOCs) using bimetallic chromium-copper supported on modified H-ZSM-5 catalyst, J. Hazard. Mater., 2006, 129(1), 39–49.
68. P. Wang, J. Wang, X. An, J. Shi, W. Shangguan, X. Hao, G. Xu, B. Tang, A. Abudula and G. Guan, Generation of abundant defects in Mn-Co mixed oxides by a facile agar-gel method for highly efficient catalysis of total toluene oxidation, Appl. Catal., B, 2021, 282, 119560.
69. K. Persson, L. D. Pfefferle, W. Schwartz, A. Ersson and S. G. Järås, Stability of palladium-based catalysts during catalytic combustion of methane: The influence of water, Appl. Catal., B, 2007, 74(3), 242–250.
70. Y. J. Li, A. Sundermann, O. Gerlach, K. B. Low, C. C. J. Zhang, X. L. Zheng, H. Y. Zhu and S. Axnanda, Catalytic decomposition of N₂O on supported Rh catalysts, Catal. Today, 2020, 355, 608–619.
71. X. Zhang, D. Chen, C. Liang and B. Shen, Water poisoning and resistance in catalytic oxidation of VOCs from industrial flue gas, Fuel Process. Technol., 2025, 273, 108231.
72. R. Burch, P. K. Loader and N. A. Cruise, An investigation of the deactivation of Rh/alumina catalysts under strong oxidising conditions, Appl. Catal., A, 1996, 147(2), 375–394.
73. Y. W. You, Y. J. Kim, J. H. Lee, M. W. Arshad, S. K. Kim, S. M. Kim, H. Lee, L. T. Thompson and I. Heo, Unraveling the origin of extraordinary lean NO_x reduction by CO over Ir-Ru bimetallic catalyst at low temperature, Appl. Catal., B, 2021, 280, 119374.
74. C. Hahn, M. Endisch, F. J. P. Schott and S. Kureti, Kinetic modelling of the NO_x reduction by H₂ on Pt/WO₃/ZrO₂ catalyst in excess of O₂, Appl. Catal., B, 2015, 168–169, 429–440.
75. M. Haneda, Y. Tomida, T. Takahashi, Y. Azuma and T. Fujimoto, Three-way catalytic performance and change in the valence state of Rh in Y- and Pr-doped Rh/ZrO₂ under lean/rich perturbation conditions, Catal. Commun., 2017, 90, 1–4.
76. U. Lassi, R. Polvinen, S. Suhonen, K. Kallinen, A. Savimäki, M. Härkönen, M. Valden and R. L. Keiski, Effect of ageing atmosphere on the deactivation of Pd/Rh automotive exhaust gas catalysts: Catalytic activity and XPS studies, Appl. Catal., A, 2004, 263(2), 241–248.
77. X. M. Cheng, D. P. Zhao, Y. N. Zhao, F. S. Li, S. Y. Chang, Y. K. Zhao, D. Tian, D. X. Yang, K. Z. Li and H. Wang, Thermal deactivation mechanism of the commercial Pt/Pd/Rh/CeO₂-ZrO₂/Al₂O₃ catalysts aged under different conditions for the aftertreatment of CNG-fueled vehicle exhaust, Fuel, 2023, 344, 128009.
78. H. Nakayama, M. Nagata, H. Abe and Y. Shimizu, In situ TEM study of Rh particle sintering for three-way catalysts in high temperatures, Catalysts, 2021, 11(1), 19.
79. E. Varga, P. Pusztai, A. Oszkó, K. Baán, A. Erdőhelyi, Z. Kónya and J. Kiss, Stability and temperature-induced agglomeration of Rh nanoparticles supported by CeO₂, Langmuir, 2016, 32(11), 2761–2770.
80. G. Goula, G. Botzolaki, A. Osatiashtiani, C. M. A. Parlett, G. Kyriakou, R. M. Lambert and I. V. Yentekakis, Oxidative thermal sintering and redispersion of Rh nanoparticles on supports with high oxygen ion lability, Catalysts, 2019, 9(6), 541.
81. K. Lee, H. Kosaka, S. Sato, T. Yokoi and B. Choi, Effect of Cu content and zeolite framework of n-C₄H₁₀-SCR catalysts on de-NO_x performances, Chem. Eng. Sci., 2019, 203, 28–42.
82. M. Machida, H. Yoshida, N. Kamiuchi, Y. Fujino, T. Miki, M. Haneda, Y. Tsurunari, S. Iwashita, R. Ohta, H. Yoshida, J. Ohyama and M. Tsushida, Thermal aging of Rh/ZrO₂-CeO₂ three-way catalysts under dynamic lean/rich perturbation accelerates deactivation via an encapsulation mechanism, ACS Catal., 2023, 13(6), 3806–3814.
83. E. A. Alikin and A. A. Vedyagin, High temperature interaction of rhodium with oxygen storage component in three-way catalysts, Top. Catal., 2016, 59, 1033–1038.
84. G. Cheng, G. Shen, J. Wang, Y. Wang, W. Zhang, J. Wang and M. Shen, The hydrothermal stability and the properties of non- and strongly-interacting Rh species over Rh/γ, θ-Al₂O₃ catalysts, Catalysts, 2021, 11(1), 99.
85. V. O. Stoyanovskii, A. A. Vedyagin, G. I. Aleshina, A. M. Volodin and A. S. Noskov, Characterization of Rh/Al₂O₃ catalysts after calcination at high temperatures under oxidizing conditions by luminescence spectroscopy and catalytic hydrogenolysis, Appl. Catal., B, 2009, 90(1–2), 141–146.
86. K. Shimura and T. Fujitani, Effects of rhodium catalyst support and particle size on dry reforming of methane at moderate temperatures, Mol. Catal., 2021, 509, 111623.
87. M. Ghelamallah and P. Granger, Impact of barium and lanthanum incorporation to supported Pt and Rh on α-Al₂O₃ in the dry reforming of methane, Fuel, 2012, 97, 269–276.
88. H. C. Yao, H. K. Stepien and H. S. Gandhi, Metal-support interaction in automotive exhaust catalysts: Rh-washcoat interaction, J. Catal., 1980, 61, 547–550.
89. J. J. Thiart, H. J. Viljoen, K. Kriel, J. Puszynski, J. E. Gatica and V. Hlavacek, Development of thermal stresses in reacting media-I. Failure of catalyst particle, Chem. Eng. Sci., 1991, 46(1), 351–359.
90. P. Kompio, A. Brückner, F. Hipler, O. Manoylova, G. Auer, G. Mestl and W. Grünert, V₂O₅-WO₃/TiO₂ catalysts under thermal stress: Responses of structure and catalytic behavior in the selective catalytic reduction of NO by NH₃, Appl. Catal., B, 2017, 217, 365–377.
91. R. Jiang, B. Yi, Q. Wei, Z. He, Z. Sun, J. Yang and W. Hua, Study on the mechanism of carbon nanotube-like carbon deposition in tar catalytic reforming over Ni-based catalysts, J. Environ. Manage., 2024, 362, 121349.
92. W. K. Jóźwiak, M. Nowosielska and J. Rynkowski, Reforming of methane with carbon dioxide over supported bimetallic catalysts containing Ni and noble metal: I. Characterization and activity of SiO₂ supported Ni-Rh catalysts, Appl. Catal., A, 2005, 280(2), 233–244.
93. C. Liu, L. Chen, J. Li, L. Ma, H. Arandiyan, Y. Du, J. Xu and J. Hao, Enhancement of activity and sulfur resistance of CeO₂ supported on TiO₂-SiO₂ for the selective catalytic reduction of NO by NH₃, Environ. Sci. Technol., 2012, 46(11), 6182–6189.
94. B. Yang, Y. Shen, S. Shen and S. Zhu, Regeneration of the deactivated TiO₂-ZrO₂-CeO₂/ATS catalyst for NH₃-SCR of NO_x in glass furnace, J. Rare Earths, 2013, 31(2), 130–136.
95. Y. Yu, C. He, J. Chen, L. Yin, T. Qiu and X. Meng, Regeneration of deactivated commercial SCR catalyst by alkali washing, Catal. Commun., 2013, 39, 78–81.
96. J. Wu, J. Chen, Z. B. Ding, H. Yang, X. Wang and Z. Rui, Strategies for improving sulfur resistance of lean methane oxidation catalysts: Progresses and future perspectives, Chem. Eng. J., 2024, 479, 147640.
97. A. Almofleh and H. A. Aljama, Boron doping to limit sulfur poisoning on metal catalysts, ChemCatChem, 2023, 15(7), e202201545.
98. S. Q. Feng, M. Jiang, X. E. Song, P. Z. Qiao, L. Yan, Y. T. Cai, B. Li, C. Y. Li, L. L. Ning, S. Y. Liu, W. Q. Zhang, G. R. Wu, J. Y. Yang, W. R. Dong, X. M. Yang, Z. Jiang and Y. J. Ding, Sulfur poisoning and self-recovery of single-site Rh₁/Porous organic polymer catalysts for olefin hydroformylation, Angew. Chem., Int. Ed., 2023, 62(30), e202304282.
99. J. Hu, Y. Zhai, S. C. Li, L. Li, F. S. Tang, L. Ruan and Z. Zhang, Improvement of NO adsorptive selectivity by the embedding of Rh in MOF-177 as carrier, Appl. Organomet. Chem., 2023, 37(4), e7051.
100. Y. Li, K. B. Low, A. Sundermann, H. Zhu, L. E. Betancourt, C. Kang, S. Johnson, S. Xie and F. Liu, Understanding the nature of Pt-Rh synergy for three-way conversion catalysis, Appl. Catal., B, 2023, 334, 122821.
101. D. Chen, W. Zhao, Z. Xu, Z. Zhao, J. Yang, Y. Hou, Y. Zhang, Z. Feng, M. Cui and X. Huang, The structure-activity relationships of Rh/CeO₂-ZrO₂ catalysts based on Rh metal size effect in the three-way catalytic reactions, Nano Res., 2024, 17(8), 6870–6878.
102. Y. Zhang, P. Glarborg, K. Johansen, M. P. Andersson, T. K. Torp, A. D. Jensen and J. M. Christensen, A rhodium-based methane oxidation catalyst with high tolerance to H₂O and SO₂, ACS Catal., 2020, 10(3), 1821–1827.
103. Q. Sun, B. Aguila, G. Verma, X. Liu, Z. Dai, F. Deng, X. Meng, F.-S. Xiao and S. Ma, Superhydrophobicity: Constructing homogeneous catalysts into superhydrophobic porous frameworks to protect them from hydrolytic degradation, Chem, 2016, 1(4), 628–639.
104. Y. S. Chen, J. Fan, J. Deng, X. Jiang, Y. Jiao and Y. Q. Chen, Synthesis of high stability nanosized Rh/CeO₂-ZrO₂ three-way automotive catalysts by Rh chemical state regulation, J. Energy Inst., 2020, 93(6), 2325–2333.
105. H. Kawabata, Y. Koda, H. Sumida, M. Shigetsu, A. Takami and K. Inumaru, Active three-way catalysis of rhodium particles with a low oxidation state maintained under an oxidative atmosphere on a La-containing ZrO₂ support, Chem. Commun., 2013, 49(38), 4015–4017.
106. M. Haneda, Y. Tomida, H. Sawada and M. Hattori, Effect of rare earth additives on the catalytic performance of Rh/ZrO₂ three-way catalyst, Top. Catal., 2016, 59(10), 1059–1064.
107. M. Li, X. Wu, Y. Cao, S. Liu, D. Weng and R. Ran, NO reduction by CO over Rh/Al₂O₃ and Rh/AlPO₄ catalysts: Metal-support interaction and thermal aging, J. Colloid Interface Sci., 2013, 408, 157–163.
108. E. A. Alikin, S. P. Denisov, K. V. Bubnov and A. A. Vedyagin, Self-regeneration effect of three-way catalysts during thermal aging procedure, Catalysts, 2020, 10(11), 1257.
109. A. A. Vedyagin, V. O. Stoyanovskii, R. M. Kenzhin, E. M. Slavinskaya, P. E. Plyusnin and Y. V. Shubin, Purification of gasoline exhaust gases using bimetallic Pd-Rh/δ-Al2O3 catalysts, React. Kinet., Mech. Catal., 2019, 127(1), 137–148.
110. M. Machida, H. Yoshida, N. Kamiuchi, Y. Fujino, T. Miki, M. Haneda, Y. Tsurunari, S. Iwashita, R. Ohta, H. Yoshida and J. Ohyama, Rh nanoparticles dispersed on ZrO₂-CeO₂ migrate to Al₂O₃ supports to mitigate thermal deactivation via encapsulation, ACS Appl. Nano Mater., 2023, 6(11), 9805–9815.
111. A. Jalal Samed, U. Kosuke, I. Keita and M. Machida, Catalytic NO-CO-C₃H₆-O₂ reactions over hydrothermally stable Rh-supported ZrP₂O₇ catalyst, Int. J. Environ. Stud., 2013, 70(4), 549–559.
112. S. Misumi, H. Yoshida, S. Hinokuma, T. Sato and M. Machida, A nanometric Rh overlayer on a metal foil surface as a highly efficient three-way catalyst, Sci. Rep., 2016, 6, 29737.
113. E. A. Alikin, S. P. Denisov and A. A. Vedyagin, Partial regeneration of model TWC after high-temperature aging on engine bench, Top. Catal., 2019, 62, 324–330.
114. G. Lin, S. Cheng, S. Wang, T. Hu, J. Peng, H. Xia, F. Jiang, S. Li and L. Zhang, Process optimization of spent catalyst regeneration under microwave and ultrasonic spray-assisted, Catal. Today, 2018, 318, 191–198.
115. S. Rossignol and C. Kappenstein, Effect of doping elements on the thermal stability of transition alumina, Int. J. Inorg. Mater., 2001, 3(1), 51–58.
116. E. A. Alikin, E. O. Baksheev, G. B. Veselov, R. M. Kenzhin, V. O. Stoyanovskii, P. E. Plyusnin, Y. V. Shubin and A. A. Vedyagin, Advantages and disadvantages of barium oxide addition to bimetallic Pd-Rh three-way catalysts supported on zirconia-doped alumina, J. Environ. Chem. Eng., 2024, 12(3), 112945.
117. V. O. Stoyanovskii, A. A. Vedyagin, A. M. Volodin, R. M. Kenzhin, Y. N. Bespalko, P. E. Plyusnin and Y. V. Shubin, Optical spectroscopy of Rh³⁺ ions in the lanthanum-aluminum oxide systems, J. Lumin., 2018, 204, 609–617.
118. V. O. Stoyanovskii, A. A. Vedyagin, A. M. Volodin, R. M. Kenzhin, Y. V. Shubin, P. E. Plyusnin and I. V. Mishakov, Peculiarity of Rh bulk diffusion in La-doped alumina and its impact on CO oxidation over Rh/Al₂O₃, Catal. Commun., 2017, 97, 18–22.
119. S. Jang, D. G. Oh, H. Kim, K. H. Kim, K. Khivantsev, L. Kovarik and J. H. Kwak, Controlling the phase transformation of alumina for enhanced stability and catalytic properties, Angew. Chem., Int. Ed., 2024, 63(15), e202400270.
120. P. G. W. A. Kompio, A. Brückner, F. Hipler, O. Manoylova, G. Auer, G. Mestl and W. Grünert, V₂O₅-WO₃/TiO₂ catalysts under thermal stress: Responses of structure and catalytic behavior in the selective catalytic reduction of NO by NH₃, Appl. Catal., B, 2017, 217, 365–377.
121. A. M. Beale, L.-G. Ines, M. Teuvo and R. G. Palgrave, Development and characterization of thermally stable supported V-W-TiO₂ catalysts for mobile NH₃-SCR applications, Catal., Struct. React., 2015, 1(1), 25–34.
122. T. Pellegrinelli, G. McCullough, R. Caporali, I. Murray, V. Celorrio, E. K. Gibson, C. Hardacre and A. Goguet, Thermal ageing of a commercial LNT catalyst: Effects on the structure and functionalities, Catal. Today, 2022, 384–386, 228–237.
123. T. Osaki and T. Mori, Role of potassium in carbon-free CO₂ reforming of methane on K-promoted Ni/Al₂O₃ catalysts, J. Catal., 2001, 204(1), 89–97.
124. R. G. Ding and Z. F. Yan, Structure characterization of the Co and Ni catalysts for carbon dioxide reforming of methane, Catal. Today, 2001, 68(1), 135–143.
125. X. Du, D. Zhang, L. Shi, R. Gao and J. Zhang, Coke- and sintering-resistant monolithic catalysts derived from in situ supported hydrotalcite-like films on Al wires for dry reforming of methane, Nanoscale, 2013, 5(7), 2659–2663.
126. A. M. O'Connor and J. R. H. Ross, The effect of O₂ addition on the carbon dioxide reforming of methane over Pt/ZrO₂ catalysts, Catal. Today, 1998, 46(2), 203–210.
127. W. Sun, L. C. Lin, X. Peng and B. Smit, Computational screening of porous metal-organic frameworks and zeolites for the removal of SO₂ and NO_x from flue gases, AIChE J., 2014, 60(6), 2314–2323.
128. K. Yang, F. Xue, Q. Sun, R. Yue and D. Lin, Adsorption of volatile organic compounds by metal-organic frameworks MOF-177, J. Environ. Chem. Eng., 2013, 1(4), 713–718.
129. L. Tan, L. L. Huang, Y. C. Liu and Q. Wang, Detailed mechanism of the NO plus CO reaction on Rh (100) and Rh (111): A first-principles study, Appl. Surf. Sci., 2018, 444, 276–286.
130. J. Y. Liu, L. Tan, L. L. Huang, Q. Wang and Y. C. Liu, Kinetic monte carlo modeling for the NO-CO reaction mechanism on Rh (100) and Rh (111), Langmuir, 2020, 36(12), 3127–3140.
131. Z. Luo, G. Zhao, H. Pan and W. Sun, Strong metal-support interaction in heterogeneous catalysts, Adv. Energy Mater., 2022, 12(37), 2201395.
132. M. Li, T. Zhang, S. Z. Yang, Y. Sun, J. Zhang, F. Polo-Garzon, K. M. Siniard, X. Yu, Z. Wu, D. M. Driscoll, A. S. Ivanov, H. Chen, Z. Yang and S. Dai, Mechanochemistry-induced strong metal-support interactions construction toward enhanced hydrogenation, ACS Catal., 2023, 13(9), 6114–6125.

---