Progress in Catalytic Oxidation of Noble Metal-Based Carbon Monoxide: Oxidation Mechanism, Sulfur Resistance, and Modification

Abstract

Carbon monoxide (CO) is an important air pollutant generated from the incomplete combustion of fossil fuels, particularly in industrial processes such as iron and steel smelting, power generation, and waste incineration, posing environmental challenges that demand effective removal strategies. Recent advances in noble metal catalysts for catalytic oxidation of CO, particularly Pt-, Pd-, and Rh-based systems, have been extensively studied. However, there is still a lack of systematic review on noble metal-based catalytic oxidation of CO, especially regarding the effects of different active components of the catalysts and the mechanism of sulfur resistance. Based on extensive research and literature findings, this study comprehensively concluded the advances in noble metal-based catalytic oxidation of CO. The effects of preparation methods, supports, and physicochemical properties on the catalytic performance of CO were explored. In addition, the mechanism of the catalytic oxidation of CO were further summarized. Furthermore, given the prevalence of SO₂ in the flue gas, the mechanism of sulfur poisoning deactivation of catalysts and the anti-sulfur strategies were further reviewed. Exploration of new supporting materials, catalyst surface reconstruction, doping modification, and other catalyst design strategies demonstrate potential in improving sulfur resistance and catalytic efficiency. This study provides valuable insights into the design and optimization of noble metal-based catalysts for the catalytic oxidation of CO.

1. Introduction

The incomplete combustion of carbon during fossil fuel utilization generates carbon monoxide (CO), a significant air pollutant in industrial flue gases. Industrial processes such as iron and steel production, power generation, and waste incineration emit large quantities of CO-containing flue gases, which are often released into the atmosphere without proper treatment, contributing to environmental pollution [1]. The CO emission concentration in the flue gas of power plant and coke oven is generally in the range of 100–1300 mg/m³ and 2000–4000 mg/m³, respectively [2]. The CO concentration in the sintering flue gas can even reach 10,000 mg/m³ [3]. CO is a colorless, odorless, and hazardous gas. Atmospheric concentrations exceeding 0.1% can lead to asphyxiation due to its strong affinity for hemoglobin—approximately 210 times greater than that of oxygen (O₂). This binding severely impairs oxygen transport in the bloodstream, affecting the central nervous system and causing symptoms such as dizziness, headaches, and permanent brain damage. In extreme cases, prolonged exposure can result in fatal asphyxiation [4]. In addition, CO also represents a substantial energy loss due to its high calorific value. Additionally, as one of the most reactive trace gases in the atmosphere, CO plays a critical role in atmospheric chemistry, indirectly contributing to global warming and ozone depletion [5]. Consequently, the effective removal of CO from industrial flue gases remains a pressing challenge for air pollution control and energy efficiency optimization [6,7].

Over the past two decades, mitigating CO emissions from incomplete fossil fuel combustion has emerged as a critical challenge across multiple industries [8]. Current CO treatment technologies can be broadly classified into four categories based on their underlying principles: adsorption, solvent absorption, direct combustion, and catalytic oxidation [9,10]. However, adsorption and solvent absorption methods exhibit limited applicability—adsorption is suitable only for low-concentration CO flue gases, while solvent absorption is more effective for high-concentration streams. Additionally, both technologies incur substantial operating costs. Direct combustion, though effective, requires high-temperature flue gas, rendering it unsuitable for low-temperature applications such as coke oven and sintering flue gas treatment. In contrast, catalytic oxidation technology has garnered significant attention due to its versatility in treating both high- and low-concentration CO emissions and its ability to facilitate CO oxidation to CO₂ at lower temperatures via catalytic action. As a gas-solid phase catalytic reaction, this method substantially reduces activation energy, accelerating the reaction rate. Key advantages include low ignition temperature, energy efficiency, broad applicability, and high purification effectiveness [11]. Furthermore, the simplicity of the CO oxidation process has made it a model reaction for investigating catalyst structure-activity relationships and reaction mechanisms [8]. Extensive research has focused on a variety of CO oxidation catalysts, and the noble metal-based catalysts have received widespread attention in CO catalytic purification due to their high catalytic efficiency and good water and sulfur resistance, such as Pt [12,13,14], Pd [8,15,16], and Au [7] etc. In addition, research has revealed that components of flue gas can induce catalyst poisoning. H₂O and SO₂ in the flue gas can affect the performance and stability of the catalyst [17]. Therefore, studies on reducing the noble metal loading and improving the catalyst’s resistance to poisoning are particularly important for improving CO catalytic performance. Currently, recent advances in noble metal catalysts for catalytic oxidation of CO (particularly Pt, Pd, and Rh systems) have been extensively investigated, primarily focusing on the active components of the catalysts [12,18,19], the catalytic mechanism [20], and the low-temperature catalysis [17,21]. There is still a lack of systematic review on the noble metal catalysts catalytic oxidation of CO in industrial flue gases, especially a comprehensive overview of the different active components of the catalysts and the mechanism of sulfur resistance.

Based on extensive research results and literature findings, this study comprehensively concluded the application of typical noble metal-based catalysts in the field of catalytic oxidation of CO. The effects of preparation methods, supports, and physicochemical properties on the catalytic performance of CO were explored. In addition, the mechanism of the catalytic oxidation of CO were further summarized. Furthermore, given the prevalence of SO₂ in the flue gas, the mechanism of sulfur poisoning deactivation of catalysts and the anti-sulfur strategies were further reviewed. The study done could provide guidance for the development of optimal design of noble metal-based catalysts for the catalytic oxidation of CO.

2. Noble Metal Catalysts for Catalytic Oxidation of CO

Over the last few decades, a vast number of noble metal catalysts have been created and employed for CO oxidation [5]. In the context of the catalytic oxidation of CO reaction, noble metals display excellent adsorption and activation properties towards CO and O₂. These characteristics endow noble metals with a high degree of catalytic efficiency in the complete oxidation of CO, positioning them as the catalyst of choice for this reaction. In comparison to non-noble metal and transition metal catalysts, noble metal catalysts for catalytic oxidation of CO present enhanced catalytic activity and more pronounced adaptability. Therefore, despite their expensive cost, noble metal catalysts’ superior performance makes them the first choice in many sectors, which is one of the reasons they were widely commercialized [22]. Additionally, synergistic catalytic effects between the active components of the noble metal-based CO catalysts and the supports were also found to be possible.

Table 1. Performance of different noble metal catalysts for CO oxidation.

Catalyst	Support	GHSV (h−1)/Flow Rate (mL/min)	Reactant	Reaction Temperature (°C)	CO Conversion Rate (%)	Reference
Pd	CeO2	480,000 h−1	0.45% CO + 10% O2	150	100	[9]
	TiO2			100	25
	Al2O3			150	80
Pd	FeOx	15,000 h−1	1% CO + 5% O2	100	98	[15]
Pt	TiO2	60,000 h−1	1% CO + 1% H2O + 50 ppm SO2	220	10	[25]
			1% CO without H2O and SO2	280	100
Pt	CeO2	240,000 h−1	0.2% CO + 1% O2	100	92	[12]
Pt	SnO2	32,000 h−1	5.0% CO + 15% O2	103	20	[28]
	Al2O3			147	20
Pt	TiO2	6,000,000 h−1	1% CO + 10% O2 + 20% H2O + 40 ppm SO2 + 40 ppm NO	500	54	[29]
Pt-Pd	Al2O3	120,000 h−1	1000 ppm C3H6 + 500 ppm NO + 0.15% CO + 10% O2 + 10% H2O	160	80	[30]
Rh	TiO2	720,000 h−1	1% CO + 5% O2	−50	100	[18]
Au	FeOx	60,000 h−1	20% O2 + 2% CO	240	50	[31]
Au	CeO2	/	2:1:22 CO/O2/N2	150	87.5	[32]
	MnO2	/		100	75
Pt-Rh	Al2O3	30,000	1.01% NO + 1.01% C3H6 + 1.03% CO + 0.99% SO2	250	90	[33]

summarizes some of the studies of noble metal-based CO catalysts. These findings not only serve as an essential reference for noble metal catalyst research but also give valuable technical assistance for the growth of related fields [23,24,25]. Studies have revealed that the platinum group metal (PGM) (mainly including Pt, Pd, and Rh) and Au supported catalysts showed excellent CO conversion at low temperatures, and are the focus of current research on noble metal-based CO catalysts [24,26,27]. Therefore, this study further addressed the development of PGM-based and Au-based catalysts for catalytic oxidation of CO.

PGM catalysts are predominantly used in industrial applications due to their superior catalytic efficiency and chemical stability, particularly for the purification of hazardous gases and pollutants [11,34]. Early application research on PGM catalysts concentrated on catalysts such as Pt and Pd wires or metal sheets, which have a smaller specific surface and less distribution of active components, resulting in reduced catalytic activity [35]. As a result, development of Pt, Pd, and other catalysts has progressed slowly. It was not until Engel et al. [36] examined the catalytic oxidation behavior of CO on monocrystalline Pd particles utilizing cross-molecular beam technology that high dispersion was discovered to be an essential feature in high catalytic activity. Among the PGM-based noble metal catalysts, the Pt- and Pd-based catalysts exhibit excellent CO catalytic activity at low temperatures. Therefore, the applications of Pt, Pd, and other loaded catalysts for catalytic oxidation of CO were analyzed with emphasis on exploring the effects of preparation methods, noble metal distribution states, and other physicochemical properties.

2.1. Pt-Based Catalysts for Catalytic Oxidation of CO

The Pt-based catalysts have been a hot research topic for noble metal catalysts due to their excellent low-temperature catalytic conversion efficiency for CO [37]. CO showed strong adsorption at the Pt site, which suppressed the adsorption activity of O₂ at low temperatures to some extent [38]. The preparation method plays a crucial role in the size and morphology of the active metal on the Pt-based catalysts, which determines the catalytic efficiency of the catalysts. This is mainly due to the differences in the pore structure between the interiors of catalyst particles obtained by different catalyst preparation methods, which affects the dispersion of Pt on the catalyst surface [39]. Cai et al. [40] investigated the effects of three catalyst preparation methods, including impregnation, dry ball mill, and wet ball mill, on the catalytic activity of Pt/TiO₂ catalysts for simulated CO oxidation in sintering flue gas. The simulation results showed that the Pt/TiO₂ prepared by the impregnation method exhibited better catalytic activity, due to the high dispersion of Pt on the surface of the catalyst with small particle size. Hong and Sun [41] compared the CO oxidation activity of Pt/CeO₂ catalysts prepared by impregnation, deposition-precipitation, and impregnation-reduction methods, and found that the catalyst prepared by the impregnation-reduction method performed better catalytic activity. The enhanced O₂ adsorption observed on Pt/CeO₂ catalysts prepared via the impregnation method may be linked to a higher concentration of metallic Pt(0) species present on their surface.

Pt-based catalysts with different supports showed differentiated CO oxidation performance. The Pt/SnO₂ catalyst has a lesser specific surface area than the γ-Al₂O₃ catalyst but exhibits high catalytic activity [28]. The reduction temperature of Pt species on the Pt/SnO₂ catalyst was lower than room temperature, and the reduced Pt species at lower temperatures facilitated complete CO conversion. Therefore, the catalytic efficiency of Pt-based systems in CO conversion exhibits strong dependence on structural parameters including metallic cluster dimensions, atomic dispersion, and oxidation state distribution. The Pt/TiO₂ catalyst with large dispersion and small particle size of Pt species (Pt/TiO₂ (R)-600) was observed to exhibit enhanced catalytic activity at room temperature when it was treated with H₂ [13]. In addition, a differentiated activation capacity for CO was evidenced by Pt species, with CO being more readily activated at the Pt⁰ site than at the Pt²⁺ site due to the larger Pt_metallic/Pt_total values [13]. Investigations on brookite-TiO₂-supported Pt catalyst demonstrated that enhanced CO conversion rates under identical loading conditions directly correlated with both increased Pt⁰ specific and improved nanoparticle dispersion of Pt [42,43]. This could be mainly attributed to the easier desorption of CO adsorbed on the Pt⁰ sites and its participation in the reaction, whereas CO adsorbed more strongly on the Pt^δ+ and Pt²⁺ sites [42,43]. Zheng et al. [44] discovered a Pt/MgFe₂O₄ catalyst with strong CO oxidation activity at room temperature and the structure of MgFe₂O₄ is a kind of support capable of providing uncoordinated lattice oxygen as active oxygen to participate in the CO oxidation reaction. Lee et al. [45] heated CeO₂ at 800 °C, reducing surface defects, weakening the interaction between the active component and the support, forming a small amount of Pt-O-Ce bond, improving Pt dispersion in the catalyst, and improving catalytic performance. Wang et al. [46] found that the shape of WO₃ has significant effects on the CO oxidation activity of the Pt catalyst, and the Pt catalyst supported on WO₃ nanolamella has higher CO oxidation activity. This was mainly attributed to the higher dispersion of Pt on the support and more Pt⁰ species formed during the metal-support interaction, which was more conducive to CO activation. Lou et al. [47] synthesized single-atom Pt catalysts, including Pt/Fe₂O₃, Pt/ZnO, and Pt/γ-Al₂O₃ and found that water can promote the catalytic oxidation of CO showing high catalytic activity to CO. Therefore, the CO oxidation activity of these catalysts is independent of the type of support. However, after heat treatment at 300 °C to remove the –OH group on the support, they found that the Pt catalyst on the reducible substance (Fe₂O₃) had the highest CO catalytic activity [48]. An et al. [49] discovered that when compared with pure Pt and pure oxides (such as Co₃O₄, NiO, MnO₂, Fe₂O₃, and CeO₂), the oxidation activity of the Pt/oxide catalyst on CO was greatly increased. The metal-support interface on the catalyst is critical to improving CO oxidation activity because it provides an active site for lattice oxygen [50]. In the Pt/Co₃O₄ catalyst, the interface between the metal and the support forms the CoO active site, increasing the catalyst’s activity, as illustrated in Figure 1. Song et al. [51] prepared Pt/oxide nanowires composite two-dimensional catalysts with a cross-stacked structure of Pt nanowires and oxide (CoO or TiO₂) nanowires (see Figure 2). It was found that Pt/CoO NW had significantly better catalytic performance than Pt/TiO₂ NW or Pt NW on the SiO₂ substrate, which was attributed to the support effect and low bonding energy of metal atoms (Co and Ti) with oxygen atoms of supported metal oxide nanowires [52]. These studies discovered that the catalyst support not only keeps the catalyst stable, but it also controls the active component’s dispersion and reactivity. Furthermore, the physical and chemical features of the catalyst support can influence the performance of the CO oxidation reaction. As a result, selecting and optimizing the catalytic support is crucial in the development of an efficient catalyst.

2.2. Pd-Based Catalysts for Catalytic Oxidation of CO

The Pd-based catalysts exhibit good thermal stability and commercial viability. The property of support is an important factor affecting the performance of Pd-based catalyst. Transition metal oxides (e.g., TiO₂, MnOx, CeO₂) are commonly used as support for noble metals such as Pd during CO oxidation due to their outstanding O₂ storage and redox capacity [15]. Kochubey et al. [53,54] developed a range of Pd catalysts using SiO₂, TiO₂, and Al₂O₃ as carriers, and investigated the structure of Pd particles as well as the adsorption properties of CO and O₂. The results showed that the Pd on SiO₂ has a rather perfect structure, with the Pd-Pd distance in the bulk close to that of a Pd foil. This was mainly due to the uniform distribution of Pd particles on SiO₂, with the largest size indicating that Pd(OAc)₂ was uniformly adsorbed on the SiO₂ after impregnation. However, the Pd on TiO₂ and γ-Al₂O₃ were flattened Pd clusters, which was mainly caused by the interaction of Pd with the support strongly interfering with the structure of the Pd clusters. Satsuma et al. [9] investigated the catalytic performance of Pd loaded on CeO₂, TiO₂, Al₂O₃, ZrO₂, and SiO₂ under low-temperature CO oxidation conditions. The results showed that CeO₂ tended to exhibit higher activity with sufficient oxygen storage capacity at 100 °C (2.1 mmol/g). This could be mainly because the presence of Ce⁴⁺/Ce³⁺ ion pairs on the CeO₂ support promoted the generation of oxygen vacancies, which effectively adsorbed and released the O₂. In addition, a significant reduction in the initial temperature required for CO oxidation by approximately 130 °C was found with the introduction of CeO₂ into Pd/Al₂O₃ catalysts [55]. The introduction of CeO₂ created defects at the Pd/Ce interface, which facilitated the activation and oxidation of CO and O₂ at the Pd/Ce interface. Studies also showed that Ba and La doping significantly increased the catalytic activity of Pd/Al₂O₃ for CO oxidation, due to the improved dispersion of Pd on the catalyst surface and favorable electron modification ratio, i.e., higher Pd²⁺/Pd-0 [55]. It was proved that the size of Pd also has a significant effect on the catalyst activity. Wang et al. [16] employed PdCl₂, [Pd(NH₃)₄](NO₃)₂, and Pd(acac)₂ as precursors to synthesize Pd catalysts with varied size distributions, which were then characterized to detect Pd dispersion. The results demonstrated that the catalytic activity of CO increased with decreasing Pd particle size and increasing the O₂/CO ratio. In particular, the Pd/TiO₂ catalyst produced with Pd(acac)₂ as the precursor has highly dispersed Pd particles, which produce strong catalytic oxidation of CO activity at near room temperature. Choudhary et al. [56] discovered that the activity of supported Pd-based catalysts is determined not only by the amount of PdO present, but also by the degree of oxidation and Pd presence. When the PdO/Pd ratio is equivalent and PdO is exposed to the outer layer of PdO/Pd, the catalyst activity improves [56]. The shape of PdO, the exposed crystal surface, and the pretreatment conditions of the support all have a significant impact on the activity of the Pd catalyst [57,58].

2.3. Pt-Pd Based Catalysts for Catalytic Oxidation of CO

By combining the advantages of different metals, bimetallic catalysts can exhibit better performance than single metal catalysts. Pt-Pd bimetallic catalysts exhibit greater effectiveness in contrast to monometallic catalysts consisting solely of Pd or Pt [30]. The increase in efficiency results from the simultaneous presence of Pt and Pd active sites, which stimulates the high catalytic activity. It has also been noted that certain features lacking in monometallic Pt and Pd catalysts are present in bimetallic Pt-Pd catalysts, i.e., better temperature and water stability, and higher catalytic activity. Therefore, Pt-Pd catalyst has been widely studied because of its catalytic potential in CO oxidation. Sheintuch and Schryer [59,60] produced a series of Pd/SnO₂ and Pt/SnO₂ supported catalysts and discovered that their activity was much higher than that of metal Pt and Pd due to the high dispersion of Pt and Pd and the overflow of reactive species (such as CO and O₂) on the support. Grasping how the ratios of Pt to Pd affect the performance of Pt-Pd bimetallic catalysts holds great significance in the process of CO oxidation. Daneshvar et al. [61] developed a global kinetic model for the oxidation of a mixture containing CO, C₂H₄, C₇H₈, C₆H₁₄, and C₂H₆ over Pt-Pd/γ-Al₂O₃ diesel oxidation catalysts with varying Pt/Pd molar ratios (100:0, 75:25, 50:50, 25:75, 0:100). They determined that, among the conditions studied, the Pt/Pd molar ratio of 1:1 yielded the most optimal catalyst. Kang et al. [62] examined the effect of the Pt/Pd ratio on CO and hydrocarbon oxidation and found that the degree of CO inhibition was a function of the Pt/Pd ratio. Specifically, the inhibitory effect of CO diminished with increasing Pd content, whereas its intrinsic activity on olefin oxidation became significant with increasing Pt content.

2.4. Au-Based Catalysts for Catalytic Oxidation of CO

Compared to other noble metals, Au has a low melting point (1063 °C), making it extremely difficult to manufacture highly active catalysts by scattering them on a support using typical methods. Therefore, previous studies believed that Au, as a catalyst, exhibited performance inferior to other noble metals. Until 1989, researchers throughout believed that Au did not possess effective catalytic properties due to its bulk phase chemical inertness [19]. Haruta et al. [63] reported that Au with a size <10 nm was disseminated over transition metal oxides by co-precipitation and showed good CO oxidation activity at −70 °C. This discovery broke this idea of Au’s inactivity. The Au catalyst is particularly active in catalytic oxidation of CO and has good catalytic performance at room temperature or even lower temperatures [64]. Furthermore, with Zhang et al. [65] revealing the active properties of Au nanoparticles in non-homogeneous phase catalytic reactions, this discovery quickly stimulated a wide range of attention in the research field.

Numerous research efforts have explored various synthesis techniques for Au-based catalysts, demonstrating their significant influence on the material’s properties. Coprecipitation, impregnation, and deposition-precipitation are the most commonly used, and they often result in quite different catalysts. Therein, the deposition–precipitation method is widely employed to synthesize monodispersed Au nanoparticles, achieving superior catalytic efficiency through strengthened interactions between the metal and its support [19,24,66]. Si et al. [7] preprepared the Au/TiO₂ catalyst supported on the obtained TiO₂ powder by the deposition–precipitation method, and based on the experimental results guessed that it was the physically adsorbed O₂ molecules rather than the chemisorbed O₂ that reacted directly with CO adsorbed on the Au sites. The preferable pH and the temperature of calcination for the deposition–precipitation method were 8~9 and 200 °C, respectively, which was slightly dependent on the nature of the support [67]. Gasior et al. [68] found that Au particles created using the sedimentation approach were in hemispherical contact with the TiO₂ support, but Au particles prepared using the immersion and photochemical deposition methods were in spherical contact. In terms of CO oxidation activity, the Au/TiO₂ catalyst created using the deposition-precipitation method is clearly superior to other preparation methods [69]. Using Au/TiO₂ as a model catalyst, the Au activity showed an increasing and then decreasing relationship with increasing particle size [70].

In addition, the properties of Au-based noble metal catalysts varied significantly according to the characteristics of the supports, due to the distinct Au particle sizes and contents. Cruz et al. [32] prepared CeO₂, MnO₂,TiO₂, and SiO₂ with nominal content of 1 wt% of Au by a modified deposition–precipitation method. The results revealed that Au in the Au/CeO₂ was amorphous and has a good dispersion on the ceria support, and hetero-sized AuNPs were evidenced on SiO₂ (10–30 nm), whereas a narrow size distribution of AuNPs was found for Au/TiO₂. Wolf and Schüth [67] prepared four supported Au catalysts including Au/TiO₂, Au/Co₃O₄, Au/Al₂O₃, and Au/ZrO₂ catalysts, and found that Au/TiO₂ and Au/Al₂O₃ had similar Au particle sizes but Au/TiO₂ was more active than Au/Al₂O₃. The catalytic performance of Au catalysts extends beyond mere particle size effects, with the metal oxide playing a crucial role that surpasses simply stabilizing the Au particles [67]. Gasior et al. [68] investigated and compared the catalytic oxidation of CO performance of catalysts made by dispersing 1 wt% Au particles on the support of main group metal oxides (MGO) (Mg, Si, Sn) and transition metal oxides (TMO) (Fe, Ti, Ce). The results showed that the activity of the TMO-supported Au catalyst was much higher than that of the MGO-supported Au catalyst. As research into Au catalysts progresses, the substitution of Au for Pd, Pt, and other noble metals as active components of CO oxidation catalysts have been a research priority. However, the features of Au nanoparticles, such as their ease of aggregation and deactivation during the catalytic process, have become factors restricting their widespread application, as well as key problems for Au-based catalysts.

2.5. Other Noble Metal Catalysts

2.5.1. Rh-Based Noble Metal Catalysts

The noble metal of Rh has long been of interest for the oxidation of CO in exhaust gases, and it was one of the first noble metals to be widely used in CO catalytic processes. Rh particles exhibit high dispersion on the CeO₂ surface, and calcination of Rh/CeO₂ effectively inhibits the aggregation of CeO₂ [71]. The three-way catalytic performance tests revealed that Rh/CeO₂ synthesized through supercritical impregnation exhibited enhanced CO oxidation activity, demonstrating the effectiveness of the supercritical CO₂ impregnation method in improving oxygen storage capacity and overall three-way catalytic performance. The sub-nanometer Rh/TiO₂ catalyst showed extraordinarily high CO oxidation activity at −50 °C, and the CO conversion temperature was lowered by roughly 100 °C compared to the published Rh-based catalyst [18]. This created sub-nanometer Rh/TiO₂ catalyst in which Rh species were widely dispersed at 0.4–0.8 nm. The conversion frequency was at least three orders of magnitude higher than that of the rhodium-based catalyst. The catalyst’s excellent performance is due to the presence of Rh clusters in a specific size range in the form of Rh-O-O-Ti (superoxide), which is required for the activation of O₂ at the Rh-TiO₂ interface, and this superoxide can react with CO adsorbed at the TiO₂ site at low temperatures. This discovery rewrites the rules of CO oxidation on PGM metal catalysts and gives critical theoretical direction for the development of highly active PGM catalysts. Similar to other noble metals, the catalytic activity of Rh catalysts could be effectively improved by doping with rare earth metals or alkali metal elements. It was found that doping CeO₂ on Rh/γ-Al₂O₃ catalyst could promote CO oxidation [23]. The introduction of CeO₂ can increase the catalyst oxygenation and reduction rate, store oxygen in the oxygen-rich phase, and release oxygen in the oxygen-poor phase through the transfer of lattice oxygen from the bulk phase to the surface. In addition, the interaction between CeO₂ and Rh can change the oxidation state of Rh-loaded particles, and there is a significant difference in the CO-O₂ kinetics observed on the surface sites (zero-valent and oxidized) of Rh catalysts before and after CeO₂ doping [72]. Although the high performance of Rh has been favored, its high price and toxic inactivation during catalysis still limit its application prospects. With the advancement of new technologies such as core-shell catalysts and alloy catalysts, it will be helpful to improve catalytic efficiency, reduce the number of noble metals, and provide new research direction to realize the practical application.

2.5.2. Ag-Based Noble Metal Catalysts

Compared with Au and Pd, Ag-based catalysts have been widely studied for their ability to absorb and activate oxygen molecules in CO-catalyzed oxygenation reactions. However, due to the low melting point of Ag (about 962 °C), the Tammann temperature is only 344 °C, which means that Ag nanoparticles are susceptible to agglomeration by the reaction temperature, resulting in catalyst deactivation [73]. By increasing the loading amount of Ag nanoparticles, the operating temperature of the catalyst could be lowered, and the catalytic efficiency could be improved; however, the increase of the loading amount would lead to the consequent increase of the cost, which is not conducive to the industrial application. Mytareva et al. [74] prepared a series of Ag/Al₂O₃ catalysts with Ag loadings ranging from 0.5 to 10% by incipient wetness impregnation. The results revealed that at low Ag loadings of 0.5 and 1 wt% Ag, the isolated Ag+ was mainly present on the Al₂O₃ surface, and the formation of Ag-n clusters started when the Ag content was increased up to 3 wt%, and higher Ag loadings (10 wt%) led to the aggregation of Ag-n clusters and the formation of relatively large Ag nanoparticles (Ag-NPS). To limit the agglomeration of Ag nanoparticles, researchers have constructed various nanostructures by constructing spatial barriers or adding catalytic additives. Zhang et al. [73] prepared the Ag catalysts using the impregnation method, followed by synthesizing Al-SBA-15 supports using Al(NO₃)₃·9H₂O as alumina source through the “pH-adjusting” method. Their findings demonstrated that incorporating a specific amount of Al (Al-SBA-15 (200-50)) enhanced the catalytic performance.

3. Mechanism of Catalytic Oxidation of CO

Extensive research has been conducted to date on the catalytic oxidation mechanisms of metal/oxide CO catalysts in dry reaction atmospheres (CO + O₂). In supported metal catalyst systems, the adsorption and activation of O₂ are generally considered the rate-limiting steps in catalytic oxidation of CO [75]. Current understanding identifies four primary reaction mechanisms governing catalytic oxidation of CO: (1) the Mars-van Krevelen mechanism (MvK mechanism), (2) the Langmuir-Hinshelwood mechanism (L-H mechanism), (3) the Eley-Rideal mechanism (ER mechanism), and (4) the Termolecular Eley-Rideal Mechanism (TER mechanism).

The MvK mechanism has been identified as the dominant pathway for CO oxidation on Pt-based catalysts with enhanced oxygen storage capacity [11,76]. In this framework: (1) CO molecules chemisorb onto metallic Pt active sites; (2) the adsorbed CO subsequently reacts with lattice oxygen from the oxide support to form gaseous CO₂, simultaneously generating oxygen vacancies in the support material; (3) molecular O₂ undergoes dissociative adsorption at these vacancy sites, completing the catalytic cycle through redox-mediated regeneration of the oxygen lattice. This substrate-mediated oxygen transfer mechanism offers two key advantages: first, it avoids competitive adsorption between CO and O₂ at metal active sites under low-temperature conditions (<150 °C); second, it optimizes the thermodynamic driving force for surface reactions. Zheng et al. [42] experimentally verified this MvK mechanism for room-temperature CO oxidation over Pt/MgFe₂O₄ catalysts, as illustrated in Figure 3.

Extensive research has elucidated the CO oxidation mechanisms on various supported Pd-based catalysts.

For Pd/Fe₃O₄, Pd/Fe₂O₃, and Pd/Co₃O₄ systems, experimental evidence confirms that catalytic oxidation of CO follows the L-H mechanism. The adsorption of CO on the Pd sites was activated, which led to reactions with the dissociated oxygen atoms [20,77]. This four-step process involves: First, CO is adsorbed and activated on the Pd sites. Second, O₂ is adsorbed and dissociates into oxygen atoms (O_ads) on the Pd sites. Third, the adsorbed CO (CO_ads) reacts with O_ads to produce CO₂. Fourth, CO₂ is desorbed. These steps can be presented by the following equations (Equations (1)–(4)).

CO_gas ↔ CO_ads

(1)

O_2gas ↔ 2O_ads

(2)

CO_ads + O_ads → CO_2ads

(3)

CO_2ads → CO_2gas

(4)

The ER mechanism proceeds through a distinct sequence of elementary steps: (1) molecular O₂ first chemisorbs on Pt active sites, (2) gas-phase CO molecules then directly react with the adsorbed oxygen species to form a carbonate-like intermediate, (3) subsequent C–O bond cleavage yields an adsorbed oxygen atom and gaseous CO₂, followed by (4) CO₂ desorption to complete the catalytic cycle [78]. Comparative studies have revealed that the ER mechanism exhibits a significantly lower activation energy barrier than the LH pathway for CO oxidation on Sn-doped carbon nanotubes (Sn-CNTs), as demonstrated through density functional theory calculations [79].

A novel approach to CO oxidation, termed the TER mechanism, has been discovered, involving a single Au atom anchored on h-BN [80]. This process involves two pre-adsorbed CO molecules that facilitate the activation of an O₂ molecule. This activation leads to the formation of an intermediate structure characterized by a pentagonal ring (see Figure 4), specifically an OCO-metal-OCO configuration. Subsequently, this intermediate dissociates, resulting in the simultaneous production of two CO₂ molecules. Despite significant advancements, key uncertainties remain regarding the activation of O₂ molecules and the role of lattice oxygen, both of which are pivotal for the catalytic oxidation of CO. Widmann and Behm [75] outlined three distinct reaction pathways for CO oxidation on Au/Oxides catalysts under anhydrous conditions (see Figure 5). In the first pathway (Figure 5A), CO adsorbs onto Au sites, while active oxygen species are generated from O₂ molecules or oxygen atoms adsorbed on Au nanoparticles. The second and third pathways (Figure 5B,C) involve reactive oxygen species located at the metal-support interface or oxygen vacancies, which react with CO to form CO₂. These processes align with the conventional L-H mechanism, as all reactants are adsorbed. In contrast, the fourth pathway (Figure 5D) involves lattice oxygen from the support acting as the active species, reacting with CO to produce CO₂, thereby adhering to the MvK mechanism.

Extensive research has demonstrated the promotional effect of water vapor on CO oxidation kinetics over Pt-based catalysts, with multiple mechanistic pathways proposed to account for this enhancement. Wang et al. [81] systematically investigated this phenomenon using single-atom Pt₁/CeO₂ catalysts, revealing that water vapor actively participates in the CO oxidation process. Their isotopic tracing experiments showed that when the reaction temperature exceeded 90 °C, approximately 50% of the produced CO₂ contained CO¹⁸O from H₂¹⁸O, providing direct evidence for water’s involvement in the catalytic cycle. Complementing these findings, Feng et al. [25] observed a significant improvement in CO oxidation performance for 0.1Pt/TiO₂-A catalysts in the presence of H₂O. Notably, their studies demonstrated that water’s promotional effect could even counteract the poisoning influence of SO₂ under co-existence conditions. Through sophisticated H₂¹⁸O isotopic labeling experiments, they elucidated that water modifies the CO oxidation pathway, leading to the development of an interfacial reaction mechanism at Pt-TiO₂ interface, as seen in Figure 6.

4. Mechanism of Sulfur Poisoning Deactivation of Catalysts

In the real catalytic reaction process, the catalyst gradually deactivates, and the deactivation rate of a catalyst with strong stability is slow, but it occurs quickly under extreme conditions. Along with the catalytic reaction process, the deactivation process includes both chemical and physical processes. Although deactivation is unavoidable, actions can be done to slow the deactivation of the catalyst. The key to industrial catalyst applications is to increase catalyst stability by investigating the causes of catalyst deactivation. Toxic deactivation is an important cause of catalyst deactivation, which is due to the enhanced chemical adsorption of the active site of the catalyst by impurities in the feedstock gas, resulting in the loss of catalyst activity. Sulfur poisoning represents a predominant toxic deactivation mode in oxidation catalysts [82,83,84,85]. Even at low concentrations, sulfur-containing substances can significantly deactivate the catalyst because they are adsorbed on the catalyst’s active site and react with it, preventing reactant adsorption or changing the adsorption capacity of the active site to other substances via electronic effects, thereby changing the catalyst’s catalytic performance.

Catalyst poisoning can be classified as selective or non-selective, and the two types of poisoning have distinct mechanisms. Non-selective poisoning occurs regardless of the qualities of the catalyst surface sites, and the poison undergoes uniform chemisorption, resulting in a linear connection between catalyst deactivation activity and poison chemisorption. Selective poisoning, on the other hand, is influenced by the active site’s properties (such as acid strength), so that the active site with the strongest adsorption properties is preferentially affected by poisoning, potentially leading to a more complex relationship between catalyst activity and the amount of poison adsorbed [76]. Sulfur-containing compounds (such as H₂S, SO₂, sulfur-containing VOCs, etc.) are poisons of most metals or metal oxide catalysts. Even at very low concentrations, sulfur species interact strongly with metals, resulting in strong metal-S bonds that limit adsorption and dissociation of reactants at the active site. SO₂ adsorption could cause the Pt(111) surface to recrystallize into Pt(100) via surface diffusion, resulting in a decrease in surface free energy [86]. The adsorption of sulfur compounds on metal catalysts not only restricted the active site but also altered the metals’ structural and electrical properties [87]. Due to its electronegativity, the sulfide can diminish the electron density of the Pt cluster, reducing Pt’s adsorption action on other reactants [88]. The sulfur resistance of the catalyst is closely related to whether the supports can be sulfated, because acidic SO₂ gas cannot be adsorbed on a support that is not easily sulfated (such as SiO₂) [27,89]. Since SiO₂ cannot store sulfur species, it allows for the quick migration of SO₂ to Pt throughout the reaction process, followed by oxidation to SO₃ and desorption, resulting in a high SO₂ oxidation reactivity. Simultaneously, some of the catalyst’s active sites sulfated rapidly. Other studies have found that sulfates formed on the precious metal catalyst support material are the cause of catalyst sulfur poisoning. For example, in the reaction of a sulfur-containing atmosphere, sulfur species combine with the catalyst support to form sulfate species, which cover the active site on the catalyst and clog the catalyst pore, resulting in catalyst deactivation [29,90].

It has also been found that in an SO₂-containing environment, the surface sulfate formed on the catalyst surface reacts differently. Sulfide causes the reduction and aggregation of Pt metal particles on the supported Pt catalyst, resulting in bigger Pt particles affecting catalytic performance. For example, Lee et al. [91] found that the presence of SO₂ would lead to the increase of Pt/Al₂O₃ catalyst Pt size, and the reduction of Pt particles. The purification of sulfur-containing flue gas has always been a challenge in the treatment of industrial waste gas. Since boiler flue gas frequently contains SO₂ and H₂S, sulfate or sulfite will be formed on the catalyst’s surface during the catalytic purification process, covering the surface active site and causing the catalyst to become inactive [92]. H₂O in the flue gas could exacerbate the toxicity of SO₂ to the catalyst because the OH- decomposed by H₂O could increase the oxidation rate of SO₂ on the catalyst surface [25]. He et al. [93] found that, using the prepared Pt/TiO₂ catalysts, SO₂ competed with the reacting gas for adsorption in the low-temperature range, resulting in catalyst inactivation. At high temperatures, SO₂ reacted with O₂ and lattice oxygen in the gas phase to form sulfuric acid species, which disrupted the surface chemical structure of the Pt/TiO₂ catalyst and led to a decrease in activity. After deactivation, the catalyst can be regenerated through washing, pickling, roasting, etc. However, if it reacts with the active ingredient to produce refractory sulfate, the catalyst is permanently deactivated.

5. Strategies for Sulfur Resistance in Catalytic Oxidation of CO Catalysts

Previous studies have shown that the presence of SO₂ in the catalytic oxidation reaction atmosphere can lead to the sulfation of the supports, and the degree of sulfation of the carrier is related to the reaction temperature and support type. SO₃²⁻ and SO₄²⁻ species coexisted on the Al₂O₃ support at temperatures <200 °C, while SO₄²⁻ species dominated at temperatures >300 °C [94]. The γ-Al₂O₃ is the most used catalyst support due to its high specific surface area and thermal stability. However, in the presence of sulfur species, the support could be sulfated and the generated Al₂(SO₄)₃ could block the pore on γ-Al₂O₃, preventing the reactant from adsorbing and activating at the active metal site, resulting in catalyst deactivation [95]. TiO₂ support is more sulfur resistant than Al₂O₃ support because the sulfate created on TiO₂ is unstable at higher temperatures. The formed sulfate is easy to decompose, and it is widely utilized as a catalyst support for the SCR reaction of NOx [33].

The pore structure of the catalyst carrier may also affect its sulfur poisoning performance to some extent [96,97,98]. Taira et al. [90] prepared a range of Pt/TiO₂ catalysts with various specific surface areas, crystal types, and pore architectures and investigated their CO catalytic combustion activity and sulfur resistance in a SO₂ and H₂O atmosphere. The results suggest that the catalyst with wide pore structure demonstrates improved stability at 250–300 °C. In contrast, the specific surface area and crystal form have less impact on the catalyst’s sulfur resistance and stability. SO₂ is oxidized to generate sulfuric acid in the presence of O₂ and H₂O in the reaction atmosphere. The boiling point of sulfuric acid varies with its concentration, and a concentration of roughly 98% can approach 300 °C. At the catalytic reaction temperature, the produced sulfuric acid will be liquid on the catalyst surface. According to the Kelvin equation (see Equation (5)), sulfuric acid condensation occurs first in small pores, while big pores are resistant to sulfuric acid condensation. As a result, the sulfuric acid gathered in the pore makes it impossible for the reaction gas to reach the active site, causing the catalyst to deactivate. Therefore, when the reaction temperature is lower than 500 °C, the catalyst with a pore size more than 10 nm is acceptable for the reaction in the environment of SO₂.

$$\ln \left(p/p_0\right)=-{\displaystyle \frac{2V_L\gamma \cos \theta }{r_{p}RT}}$$

(5)

where p/p₀ is the relative pressure; V_L is the molar liquid volume; γ is the surface tension; θ is the contact Angle; r_p is the pore radius; R is the gas constant; T is the temperature.

Surface reconstruction and doping modification are also commonly used to optimize catalyst performance. For example, Kim et al. [99] created a catalyst with Pt coated on a nano-sized cerium dioxide CeO₂-Al₂O₃ carrier with an octahedral form (see Figure 7) and investigated its CO oxidation resistance to sulfur. The catalyst containing octahedral CeO₂ had greater sulfur resistance after sulfation, and the catalytic activity was more reversible after sulfur removal. Due to the low basicity of the reconstructed nano-CeO₂, the generation of sulfate species is inhibited, providing a useful guidance for increasing the multiphase catalyst’s sulfur resistance [99].

Doping modification can also effectively increase the catalyst’s sulfur and water resistance by inhibiting SO₂ adsorption and reducing sulfate deposition, changing the reaction route, or forming strong contacts with the support and metal oxides [17,100,101,102]. For example, Xu et al. [103] synthesized a series of WO₃-modified 0.1 Pt-nW/Ti-A catalysts. Compared with 0.1 Pt/Ti-A, the 0.1 Pt-5W/Ti-A catalyst doped with 5 wt% WO₃ showed excellent CO oxidation activity in the presence of H₂O and SO₂, and the doping of WO₃ inhibited the adsorption of SO₂, while sulfur deposition on the catalyst surface did not increase with reaction duration. The addition of WO₃ significantly improved the sulfur resistance of the catalyst [103]. In addition, the addition of WS₂ to the Pt/TiO₂ catalysts could protect against SO₂ poisoning and improve CO oxidation performance [104]. Catalyst performance tests showed that there was no deactivation of the catalyst with 3.6 wt% WS₂ at 280 °C and a lower deactivation than the catalyst with 0.1 wt% WS₂ after 24 h of catalytic testing [104]. The SO₂ resistance of this catalyst with 3.6 wt% WS₂ was a result of the co-existence of WS₂ and WO₃ generated after calcination. Whereas the enhancement of CO oxidation efficiency of the catalyst with 3.6 wt% WS₂ was owing to improved Pt dispersion and increased Pt⁰/Pt_total ratio through the generation of Pt⁰. Yu et al. [105] designed a series of Mo-decorated 0.1 wt% Pt/TiO₂ catalyst (Pt/Mo/Ti). The Pt/3Mo/Ti catalyst achieved 100% CO conversion >200 °C while retaining outstanding long-term catalytic stability in the presence of SO₂ [105]. There was an interaction between Pt and Mo in 0.1Pt/3Mo/Ti, which helped to improve the efficiency of CO removal. The 0.1Pt/3Mo/Ti catalyst, with Mo deposited as atomically distributed Pt-O-Mo ensembles over the TiO₂ surface, was active for CO oxidation and highly anti-SO₂ toxicity. The improved SO₂ resistance of the catalyst could be explained by the weakening of SO₂ adsorption and strong charge transfer capacity due to the appropriate Brønsted acidity, reducing the oxidation of SO₂ to SO₃. The adsorption reactions occurred at the Mo=O sites on O=MoO₄, where SO₂ subsequently bonded with H₂O and O₂ to form readily degradable sulfuric acid analogs, thereby preventing the interaction of the active site with the carrier.

The synergistic effect of these factors improves the catalyst’s performance, and the mechanism of doping modification of the catalyst, as well as other performance improvement methods, need to be further studied and explored to meet industrial application requirements and provide a solid foundation for industrial catalyst development. Numerous studies have investigated the performance of catalysts in reaction systems containing H₂S gas. For the catalytic complete oxidation of methane, Ma et al. [106] prepared a novel sulfur-resistant SiO₂-supported Pd₄S catalyst (with a Pd load of 5 wt%) utilizing the H₂S-H₂ aqueous bubble method. The newly prepared Pd₄S/SiO₂ catalyst contains both Pd₄S and metal Pd⁰ on its surface. However, in the oxidation reaction, Pd₄S was highly stable, whereas most metal Pd⁰ was transformed into PdO. When Pd₄S coexisted with PdO, it not only protected PdO from sulfur poisoning but also had a significant impact on the catalyst’s activity. Furthermore, the concentration of H₂S in the H₂S-H₂ mixture can be adjusted to modify the Pd₄S content of the catalyst. When H₂S concentration was 7%, the Pd₄S/SiO₂ catalyst had a methane conversion rate of 96% at 400 °C and was stable even at 200 ppm H₂S [106]. Lee et al. [107] prepared the Pt/CeO₂ catalyst for the water-gas shift reaction and discovered that it exhibited steady catalytic activity in the presence of 100 ppm H₂S and high sulfur resistance even at 1000 ppm H₂S. It is worth noting that when the H₂S injection is stopped, the catalytic activity is completely restored, regardless of the H₂S concentration. The high sulfur resistance and reversibility of sulfur resistance are primarily due to the high oxygen storage capacity of the Pt/CeO₂ catalyst, which improves water-gas shift reaction reactivity and sulfur resistance while also aiding in the removal of adsorbed sulfur on Pt via an oxidation reaction with mobile oxygen generated by CeO₂. Silva et al. [108] investigated the catalytic performance of niobium-supported Pt, Au, and Cu catalysts in acidic water-gas shift reaction processes under simulated real-world circumstances. The Cu/Nb₂O₅ catalyst was inactive even in the absence of H₂S. Au/Nb₂O₅ showed excellent catalytic activity; however, it was entirely deactivated in sulfur-containing conditions. When H₂S was terminated, Au/Nb₂O₅’s catalytic activity returned, demonstrating that deactivation was reversible. In contrast, the Pt/Nb₂O₅ catalyst was highly sulfur resistant and did not deactivate even in the presence of 50 ppm and 1000 ppm H₂S [108]. Therefore, this catalyst was most suitable for use under acidic conditions. Yuan et al. [26] studied the long-term stability and sulfur resistance of SiO₂-supported Ni and NiRu catalysts in methanation reactions. The presence of H₂S molecules in the syngas caused the Ni/SiO₂ catalyst to deactivate rapidly, since sulfur adsorption limited the active site and enhances Ni⁰ particle sintering and oxidation. However, the effect of Ru on the reduction of S adsorption over the Ni catalyst depended on its geometric instead of electronic factor. When S binds to both Ni and Ru atoms at the same time, the presence of Ru could reduce its adsorption. Figure 8 depicts the adsorption of sulfur on different particles. The NiRu/SiO₂-P catalyst produced using polyethylene glycol (PEG) could promote the creation of this state. The NiRu/SiO₂-P catalyst with PEG may remain stable for 127 h in an H₂S-containing methanation process. In contrast, NiRu/SiO₂-NP and Ni/SiO₂ catalysts without PEG were less stable.

6. Conclusions

CO is an important air pollutant generated from the incomplete combustion of fossil fuels, particularly in industrial processes such as iron and steel smelting, power generation, and waste incineration, posing environmental challenges that require effective removal strategies. The research on CO oxidation catalysts has made significant strides, particularly in enhancing sulfur resistance and understanding the oxidation mechanisms involved. Based on extensive research results and literature findings, this study comprehensively concluded the advances in noble metal catalysts catalytic oxidation of CO. The conclusions are as follows: (1) The efficiency of noble metal-based catalysts for CO is significantly affected by the preparation method, support type, and the morphology and size of noble metal nanoparticles. (2) The mechanisms of catalytic oxidation of CO primarily involve the Mars-van Krevelen mechanism (MvK mechanism), Langmuir-Hinshelwood mechanism (L-H mechanism), Eley-Rideal mechanism (ER mechanism), and Termolecular Eley-Rideal Mechanism (TER mechanism). (3) Sulfate formation on the noble metal catalyst support is the primary cause of sulfur poisoning. Different catalyst design strategies show promise in improving sulfur resistance and catalytic efficiency, including the exploration of new supporting materials, catalyst surface reconstruction, doping modification, and other catalyst design strategies. Accordingly, the mechanism of catalytic oxidation of CO was comprehensively reviewed. Future research should focus on optimizing catalyst formulations and exploring new materials to further enhance performance in real-world applications, ultimately contributing to more effective CO removal strategies in industrial processes and reducing environmental pollution.