Research Progress of Pt-Based Catalysts toward Cathodic Oxygen Reduction Reactions for Proton Exchange Membrane Fuel Cells

Abstract

Proton exchange membrane fuel cells (PEMFCs) have been considered by many countries and enterprises because of their cleanness and efficiency. However, due to their high cost and low platinum utilization rate, the commercialization process of PEMFC is severely limited. The cathode catalyst layer (CCL) plays an important role in manipulating the performance and lifespan of PEMFCs, which makes them one of the most significant research focuses in this community. In the CCL, the intrinsic activity and stability of the catalysts determine the performance and lifetime of the catalyst layer. In this paper, the composition and working principle of the PEMFC and cathode catalyst layer are briefly introduced, focusing on Pt-based catalysts for oxygen reduction reactions (ORRs). The research progress of Pt-based catalysts in the past five years is particularly reviewed, mainly concentrating on the development status of emerging Pt-based catalysts which are popular in the current research field, including novel concepts like phase regulation (intermetallic alloys and high-entropy alloys), interface engineering (coupled low-Pt/Pt-free catalysts), and single-atom catalysts. Finally, the future research and development directions of Pt-based ORR catalysts are summarized and prospected.

1. Introduction

Currently, with the continuous improvement in the economic level, the speed of energy consumption is accelerated. Due to long-term over-exploitation, traditional fossil energy is drying up, and the problems of climate warming and environmental pollution caused by using them are becoming more and more serious [1,2]. These problems have become an important factor restricting development all over the world. Therefore, it is urgent to speed up the research on renewable energy technologies as alternatives [3]. Hydrogen energy has a superior energy storage density (142 MJ kg⁻¹), which can provide energy through chemical reactions directly in fuel cells and only produce water in the reaction process. Therefore, hydrogen has been widely accepted to play a pivotal role in the future global energy supply; accordingly, the large-scale application of hydrogen can effectively solve the raised problems such as the energy crisis, environmental pollution, and global warming [1,4]. According to the International Hydrogen Energy Commission, the proportion of hydrogen energy in total energy will reach 18% by 2050, and the market size will enter trillions of dollars [5,6]. In this regard, developed countries such as Europe and the United States have upgraded the utilization of hydrogen energy to a national strategy and competed to seize the highlands of the hydrogen energy industry [7,8,9]. The U.S. Department of Energy (DOE) has issued documents such as the “Department of Energy Hydrogen Program Plan” and the “US National Clean Hydrogen Strategy and Roadmap”, clarifying the important position of hydrogen energy and expecting up to 10 million tons of hydrogen production per year by 2030 [10,11].

Hydrogen fuel cell technology is an important approach to realize the wide utilization of hydrogen energy. Hydrogen fuel cells convert chemical energy (hydrogen and oxygen) into electrical energy while producing green and clean water as products, as well as heat [12,13]. Generally, hydrogen fuel cells are divided into five types, which will be briefly discussed [14]. The Alkaline Fuel Cell (AFC) with 35–45% KOH as the electrolyte has a high current density and power density, but its catalyst layers suffer from deactivation by produced carbonate blocking [15]. The Phosphoric Acid Fuel Cell (PAFC) with phosphoric acid as the electrolyte has the advantages of a low cost and high reaction rate, but the long start-up time and low waste heat recovery rate severely restrict its development [16]. The Molten Carbonate Fuel Cell (MCFC) with the mixture of two or more molten carbonates as the electrolyte can solve the problem of the low utilization rate of waste heat and has a wide selection of fuels, including methanol and hydrocarbons, in addition to hydrogen [17]. However, its electrolyte is corrosive, which severely reduces the lifespan of the fuel cell. The Solid Oxide Fuel Cell (SOFC) with composite oxide as the electrolyte can effectively avoid electrolyte corrosion and leakage and significantly increase the energy utilization to as high as 80% [18]. However, a series of problems such as its high cost, high operating temperature, and the sintering and instability of the materials pose new challenges for research and development. In comparison, the Proton Exchange Membrane Fuel Cell (PEMFC) with perfluorosulfonic acid membrane as the electrolyte is more promising as a power source for transportation due to its high energy conversion rate, low working temperature, and fast start-up speed. So far, PEMFCs have successfully been launched in the market and applied to heavy commercial vehicles, passenger cars, and other fields [19,20]. A representative example is the second-generation Toyota Mirai fuel cell, which exhibited an excellent power density (up to 5.4 kW L⁻¹), further promoting the application and development of hydrogen fuel cells [21]. Although the cost of fuel cells has significantly decreased (the current cost is about USD 45 kW⁻¹), it still falls behind the U.S. Department of Energy target (USD 30 kW⁻¹) [11]. In PEMFC systems, the cost of membrane electrode assembly (MEA) is expected to account for more than 50% of the total cost. In order to further reduce costs, it is necessary to understand the reaction process and mechanisms of MEA in PEMFCs.

The PEMFC is mainly composed of the MEA, bipolar plate (BPP), and external circuit (Figure 1a). The MEA hosts the interphases for the transport of electrons, protons, and gasses and the occurrence of electrochemical reactions such as oxygen reduction reaction (ORR) and hydrogen oxidation reaction (HOR). The performance of the MEA determines the property, life, and cost of the PEMFC. MEA is a three-in-one assembly produced by combining the proton exchange membrane (PEM), catalyst layer (CL), and gas diffusion layer (GDL) through hot pressing (Figure 1b). When the fuel cell is working, hydrogen reaches the anode side of MEA through the BPP and then diffuses through the anode GDL to the anode CL for HOR (Equation (1)). The generated proton by anodic HOR then goes through the PEM to the cathode CL, and the released electrons flow through the external circuit as a current [22]. Similarly, oxygen transfers through the cathode GDL and then adsorbs on the active sites for further ORR, together with proton and electrons (Figure 1c). Particularly for the ORR with much slower kinetics, either a four-electron reaction to produce H₂O (Equations (2) and (3)) or a two-electron reaction to produce H₂O₂ (Equations (4) and (5)) may occur at the cathode. Both of these pathways produce *OOH intermediates. According to Sabatier’s principle, the affinity of intermediates on the catalyst surface significantly affects ORR pathways and kinetics [23,24,25,26,27]. When the adsorption strength of oxygen-containing intermediates on the catalyst surface is too high, the desorption process will be inhibited, and the ORR activity of the catalyst will be low. When the intermediates are weakly bonded, the O-O bond is maintained, and H₂O₂ is the preferred reduction product [26,28,29]. In this regard, when the binding energy of the intermediate is moderate, the catalyst can exhibit higher ORR activity and promote the four-electron pathway at the cathode [13,30,31].

HOR occurs at the cathode at the anode as follows:

$$2{\mathrm{H}}_2\to 4{\mathrm{H}}^{+}+4{\mathrm{e}}^{-} \left(\mathrm{E}=0 \mathrm{vs}. \mathrm{RHE}\right)$$

(1)

ORR occurs at the cathode as follows:

$${\mathrm{O}}_2+4{\mathrm{H}}^{+}+4{\mathrm{e}}^{-}\to 2{\mathrm{H}}_2\mathrm{O} \left(\mathrm{E}=1.229 \mathrm{vs}. \mathrm{RHE}\right)$$

(2)

$${\mathrm{O}}_2+4{\mathrm{H}}^{+}+4{\mathrm{e}}^{-}\to \ast \mathrm{O}\mathrm{O}\mathrm{H}+3{\mathrm{H}}^{+}+3{\mathrm{e}}^{-}\to \ast \mathrm{O}+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to \ast \mathrm{O}\mathrm{H}+{\mathrm{H}}^{+}+{\mathrm{e}}^{-}\to 2{\mathrm{H}}_2\mathrm{O}$$

(3)

$${\mathrm{O}}_2+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to {\mathrm{H}}_2{\mathrm{O}}_2 \left(\mathrm{E}=0.670 \mathrm{vs}. \mathrm{RHE}\right)$$

(4)

$${\mathrm{O}}_2+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to \ast \mathrm{O}\mathrm{O}\mathrm{H}+{\mathrm{H}}^{+}+{\mathrm{e}}^{-}\to {\mathrm{H}}_2{\mathrm{O}}_2$$

(5)

It is mentioned above that, in the actual operation of the PEMFC, the ORR kinetics is much lower than that of HOR at the anode, so more catalysts are needed at the cathode to accelerate the reaction, which leads to an increase in the cost of the MEA and PEMFC [32,33]. On the other hand, the high working potential and local acidic environment of ORR lead to a harsh working condition for catalysts, so the ORR catalysts easily suffer from oxidation and deactivation [34,35].

As discussed above, the research and development of ORR catalysts for the cathode is mainly focused on kinetics and stability improvement. An ideal cathode ORR catalyst should promote four-electron selectivity and have a high active site density and an appropriate surface electronic structure to achieve a high charge transfer [36]. In addition, the cathode catalyst layer works under harsh conditions such as high pressure and a strong oxidative environment, so the ORR catalyst should have excellent stability, which is also the basic requirement of catalyst design [21,37]. At present, precious metals (such as Pt, Pd, etc.) are considered to be the most effective catalysts to promote the sluggish kinetics of ORR due to their unique electronic structures. Although some non-precious metal catalysts have been developed to replace Pt, their comprehensive performance still needs more effort. In this regard, increasing platinum-based catalysts’ intrinsic activity and using less platinum could still be the best approach. Conventional platinum nanoparticle-based catalysts suffer from a low platinum utilization rate and active site deactivation [38,39]. As a result, the development of platinum-based catalysts with higher platinum utilization, more excellent activity, and better stability has become one of the main research hotspots regarding PEMFCs. By controlling the particle size of the catalyst, researchers can downsize the particle to the atomic level to greatly improve the utilization and activity of the catalyst through the quantum size effect and surface effect [40]. Optimizing the surface structure of the catalyst can also effectively improve the performance of catalysts. For example, grain boundary engineering was recently proposed to induce the strain effect of undercoordinated atoms and effectively optimize the surface structure of atoms [41]. It is worth noting that the incorporation of transition metals can also reduce the amount of Pt to control the cost and improve the intrinsic activity. Therefore, the alloy catalyst obtained by integrating transition metal is one of the most representative catalysts, e.g., PtNi and PtCo [42,43,44]. Specifically, the Pt-based alloy catalyst changes the electronic structure of the Pt sites due to the coordination effect and the strain effect caused by the neighboring transition metals, thereby weakening the adsorption of oxygen-containing intermediates on the Pt surface and making it better than Pt in terms of ORR activity [45,46,47]. However, the transition metal leaching problem causes instability in catalysts. Therefore, some claimed that Pt-based alloy catalysts are not suitable for PEMFC cathode catalysts [48,49,50]. In this regard, researchers have further developed and optimized Pt-based catalysts from the perspectives of elemental composition, an ordered metal structure, and an increasing entropy value and by introducing single atoms. For example, the highly ordered structure can enhance the coordination effect and strain effect of intermetallic compounds and further weaken the adsorption of oxygen-containing intermediates on the catalyst surface [51,52,53,54]. Pt-based high-entropy alloy catalysts have been proposed, which possess four unique effects (the high-entropy effect, the lattice distortion effect, the sluggish diffusion effect, and the cocktail effect) and exhibit excellent ORR performance [55,56,57,58,59,60]. In addition, the coupling of a low-Pt catalyst and platinum group metal (PGM) -free catalyst has been emerging, both of which have abundant, stable catalytic active sites and tailorable porosity, enabling fast charge and mass transfer [61,62,63]. Furthermore, single-atom Pt catalysts can maximize atomic utilization, so some investigations introduced single Pt atoms to alter the coordination environment and preferred four-electron ORR, thereby improving the intrinsic activity of the catalyst [64,65,66,67].

Figure 1. (a) A schematic illustration of a PEM fuel cell stack. (b) The key components in an integrated MEA. (c) The working mechanisms in a PEMFC. Reprinted with permission from Ref. [34]. Copyright 2023, Springer Nature.

Figure 1. (a) A schematic illustration of a PEM fuel cell stack. (b) The key components in an integrated MEA. (c) The working mechanisms in a PEMFC. Reprinted with permission from Ref. [34]. Copyright 2023, Springer Nature.

Given by these existing progresses, this review is focused on discussing the development of emerging platinum-based ORR catalysts, primarily in the past five years. To be more specific, based on the problems existing in the application of ORR platinum-based catalysts in the cathode catalyst layer of PEMFCs, this paper reviews the recent innovative strategies adopted by researchers to improve the activity and stability of Pt-based ORR catalysts, including alloying, ordering, improving the composition and confusion degree, coupling the low-Pt catalysts and the PGM-free catalysts, single-atom catalysis, etc. Based on several examples of the most typical emerging catalysts in recent years, we critically outline the research trends and future development directions of new Pt-based catalysts. It is believed that this work can help researchers clarify the strategy of catalyst design and provide new ideas for improving the performance of catalysts and PEMFC components.

2. Pt-Based Alloy Catalysts

Pt-based alloy catalysts usually consist of platinum and transition metals (e.g., Co [68], Ni [69], Fe [70,71], Cu [72], etc.). Compared with a pure Pt catalyst, the addition of a transition metal can change the electronic structure and surface properties of PtM nanoparticles, reduce the interaction between the Pt and oxygen reactant, and then increase the intrinsic activity of the catalyst [73,74,75,76]. A study on the (111) facets of Pt and PtM catalysts through advanced surface characterization techniques revealed that the surface catalytic activity of a PtM alloy catalyst was significantly improved, and the specific activity (SA) is more than ten times that of a pure Pt catalyst (Figure 2) [47]. Similarly, Min et al. [77]. prepared carbon-supported Pt-based alloy electrocatalysts (Pt-Co, Pt-Cr, and Pt-Ni) by the incipient wetness method. The findings demonstrate that the decrease in the Pt-Pt neighboring distance is conducive to ORR throughout the alloying process. On the other hand, transition metals are abundant and inexpensive, which can reduce the amount of Pt and lessen the cost of the catalyst.

At present, researchers generally change the structure and morphology of Pt-based alloy catalysts by controlling the atomic ratio of metals or doping non-metallic atoms or non-noble transition metals [78,79,80] so as to greatly improve the ORR’s catalytic activity and stability. One of the most prevalent methods for enhancing catalyst performance is N-doping on the PtM alloy catalyst support. Zhao et al. [45] improved the performance of a PtCo/C-N catalyst (N content is 6.29%) by doping an appropriate amount of N element in the commercial PtCo alloy catalyst. The catalyst revealed a high mass activity (0.684 A mg_Pt⁻¹), which was twice that of the commercial PtCo/C catalyst (0.333 A mg_Pt⁻¹). This is because the stable metal-N bond reduces the adsorption energy of oxygenates on the Pt surface and significantly increases the desorption rate of hydroxyl groups during the ORR, thereby exhibiting good activity [76,78,79,81]. Meanwhile, the half-wave potential only decreased by 26 mV after 5000 cycles, and the loss was much lower than that of the PtCo/C catalyst (decreased by 66 mV). The presence of a Pt-N bond made the binding force between the metal and the support stronger, which was difficult to dissolve and manifested good stability [82,83]. This indicates that the doping of N element in alloy catalysts can effectively improve activity and stability. In order to further enhance the performance of the alloy catalyst, Zhu et al. [80] loaded the PtNi alloy catalyst on the N-doped hollow carbon nanorod (HCNR). The synthesized PtNi/HCNR-2 catalyst exhibited excellent electrocatalytic activity, and its mass activity (MA, 0.87 A mg_Pt⁻¹) was six times that of the Pt/C catalyst (0.15 A mg_Pt⁻¹). This is attributed to the fact that the N-doped HCNR changed the oxygen-containing adsorption energy on the surface of the PtNi alloy and promoted the ORR. Additionally, the catalyst exhibited high stability in a long-term operation because the doping of nitrogen enhanced the binding energy between PtNi and the support (HCNR) and inhibited the degradation of catalytic nanoparticles. Similarly, Xia et al. [46] prepared the PtCo@NGNS catalyst by encapsulating a Pt alloy with nitrogen-doped graphene nanosheets. The test results indicate that the electrochemical active surface area (ECSA) of the PtCo@NGNS catalyst was 76 m² g⁻¹, and more exposed active sites in the entire NGNS substrate leads to higher catalytic activity. When compared to the other catalysts, the PtCo@NGNS catalyst had a higher initial potential, a larger half-wave potential, and a faster dynamic process, demonstrating better electrocatalytic activity. In addition, N doping enhanced the metal–support interaction, which made the PtCo nanoparticles tightly bound to the NGNS substrate and improved the resistance to migration and corrosion of the catalyst. The mass activity of the PtCo@NGNS catalyst decreased significantly compared with that of the commercial Pt/C catalyst, and the stability of the PtCo@NGNS catalyst was much better than that of the Pt/C catalyst under long-term catalysis.

Compared with binary alloy catalysts, the further addition of Pd [84], Au [85], Ru [86], Rh [87], and other auxiliary metals to synthesize ternary alloy catalysts provides more possibilities for adjusting the electronic structure of the Pt surface and improving the activity and stability of Pt-based alloy catalysts. Moreover, the multi-component alloy catalyst can further decrease the toxicity of the reactive intermediates to the catalyst and maintain better stability in strong acidic or strong alkaline media [88,89]. Choi et al. [90] doped Au into a PtCo/C catalyst by gas-phase reduction and galvanic replacement. After 30,000 cycle durability tests in MEA, it was established that the current density of the gold-doped catalyst was the highest (1.63 A cm⁻²), while the current density of the commercial catalyst was the lowest (1.20 A cm⁻²). The leaching of Co in the Au-doped PtCo/C catalyst was less. According to density functional theory (DFT) calculations, in the presence of Au, the binding of oxygen on the catalyst surface is weaker, which inhibits the migration of Co atoms to the surface during the oxidation reduction reaction (Co segregation) and improves the activity and durability of the catalyst. Li et al. [91] added a transition metal M (M = Rh, Ru, and Pd) into Pt-Ni nanoflowers (NFs) to synthesize a series of tri-functional NF catalysts with abundant nanodendrites. Due to the addition of transition metals, especially Rh, the charge distribution on the Pt surface was optimized to have a uniformly dispersed morphology and outstanding ORR performance. The mass activity of the Pt₃NiRh NFs catalyst (2.17 A mg_Pt⁻¹) was 9.9 times that of the commercial Pt/C catalyst. After the durability test, the mass activity (MA) remained at 76.0%, which is 9.8 times that of Pt/C catalyst. Nie et al. [92] reported a carbon nanotube-supported PtPdCu alloy catalyst (PtPdCu/CNTs). It had excellent ORR activity (the half-wave potential was 0.888 V), which can be attributed to the addition of Pd and Cu to change the d-band center of Pt, and it produced the bi-functional effect. Furthermore, the introduction of CNTs can effectively anchor metal particles and accelerate the transmission of electrons to improve the activity and stability of the catalyst.

However, although Pt-based alloy catalysts have made remarkable progress, several issues need to be resolved before ORR catalysts are ready for the PEMFC cathode catalyst layer [93,94,95]. Even if many methods have been developed to enhance stability, the dissolution potential of transition metals in Pt alloy catalysts is much lower than that of Pt, which can lead to serious leaching problems for non-noble metals in the catalytic process [96,97,98,99,100]. Kariuki et al. [101] indicated that even a small amount of Ni transition metal leaching could lead to multiple negative effects, such as intrinsic activity reduction and the inhibition of oxygen diffusion, resulting in poisoning of the catalyst. Cui et al. [102] demonstrated that surface segregation in the octahedral Pt_xNi_1−x alloy catalyst dissolved the Ni-rich (111) surface, resulting in a sharp decrease in the activity of the alloy catalyst. Asset et al. [103] found that with the continuous leaching of Ni, the hollow nanostructure of the catalyst collapsed more seriously, which eventually led to a higher loss of ORR activity. Wang et al. [104] also believed that the Co-oxidation generated by the PtCo₃ catalyst during the catalytic process led to a decrease in catalyst performance. Fan et al. [48] discovered that in MEA tests, the dissolution of Co would also signally reduce the performance of the catalyst. In this regard, Gao et al. [105] used advanced measurement techniques such as XAS to explore the catalytic mechanism of Co and detected that the addition of Co can improve the activity of the alloy catalyst by the strain effect: when there is excessive Co (such as PtCo₃), Pt easily transforms into unstable PtO_x, thus accelerating the dissolution of the catalyst. Therefore, some researchers believe that Pt-based alloy catalysts are not suitable as ORR cathode catalysts for PEMFCs. In the face of the key problems of Pt-based alloy catalysts, new concepts regarding Pt-based catalysts, including (ordered) intermetallic catalysts, high-entropy alloy catalysts, single-atom catalysts, etc., have been intensively developed. In addition, researchers have also developed a series of new supports with strong metal–support interactions and a strong stability to cooperate with the development and application of these new Pt-based catalysts.

3. Pt-Based Intermetallic Catalysts

As discussed above, Pt-based alloy catalysts suffer from fast leaching due to harsh working conditions, including high applied potential (the high potential nature of ORR) and strong acidic media (proton accumulation for cathodic ORR), although their initial activity has been significantly improved [106,107]. Studies have shown that the proportion, arrangement, state, and position of each component metal in catalysts determine catalytic efficiency and stability [108,109,110]. Arranging metal atoms in a certain proportion and structural arrangement to transform the disordered structure into an ordered structure is a valid method to solve the above problems (Figure 3a). This new material is called an (ordered) intermetallic catalyst [111,112,113]. In comparison to Pt-based disordered alloy catalysts, ordered intermetallic catalysts have better performance in terms of stability and electrochemical activity [114,115,116]. This is because they usually have greater enthalpy of formation and a strong coupling effect when they are subjected to a high-temperature heat treatment from disorder to order, resulting in a noticeable increase in the structural stability of intermetallic compounds [117,118,119]. The metal atoms on the intermetallic catalyst are independent of each other and arranged at intervals. These structures cause a stronger coordination effect and strain effect between atoms, which can further reduce the adsorption energy of the oxygenated substance on the surface of the catalysts, so intermetallic catalysts have excellent physical and chemical properties such as high chemical stability and anti-deactivation [102,120,121]. Furthermore, in the currently reported articles, ordered intermetallic ORR catalysts with a core–shell structure for PEMFCs not only show extraordinary performance in half-cell tests (such as rotating disk electrode tests), but they also show considerable performance under MEA operating conditions [53,54,122,123].

In recent years, researchers have studied many Pt-M (M = Ni, Co, Fe, Cu, Cr, etc.) intermetallic catalysts with high order and good morphology/structure, such as cubic, square, rod, line, and frame [124,125]. The most representative structure is the core–shell structure, which usually uses Pt (or PtM alloy) as the shell and highly ordered intermetallic compounds as the core (Figure 3b) [124,126]. The advantage of this structure is that the ligand effect and surface strain effect caused by Pt as the metal shell can effectively improve the activity of the catalyst [112,116,122,127]. It has been reported that researchers have synthesized Pt-Co@Pt octahedral nanocrystals using an oil-phase method [128], which consists of ordered intermetallic compounds with Pt as the shell and the face-centered tetragonal (fct) Pt-Co intermetallic compound as the core. The data show that those crystals exhibit extremely high mass activity (2.82 A mg_Pt⁻¹) and specific activity (9.16 mA cm⁻²) as ORR catalysts, which are 13.4 times and 29.5 times that of commercial Pt/C catalysts, respectively. It is worth noting that the mass activity of the crystals decreased by only 21% after a 30,000-cycle accelerated durability test. The test results show that the core–shell structure not only presented excellent ORR activity, but also had significantly improved stability. The author believed that the excellent stability was due to the anisotropic strain and the largest active sites on the Pt shell surface, which effectively restrained the dissolution of the core transition metal. Some researchers have also confirmed the view that a Pt shell can effectively prevent the dissolution of the inner core and thus improve the stability of the catalyst through the three-dimensional tomography technique. For example, Gong et al. [52] reported a porous PdFe intermetallic compound modified by an incompletely covered Pt shell. Specifically, after directly decomposing the acetate salt precursors and controlling the annealing temperature to obtain a surface-clean and highly ordered core PdFe, the Pt shell was prepared by a mild surface process strategy, and finally, the intermetallic compound (p-o-PdFe@Pt) was obtained. This crystal also had excellent ORR activity and stability. The results of the tomographic reconstructions and co-stripe experiments indicate that the thickness of the Pt shell decreased after the accelerated durability test, the Pt atoms were more uniformly dispersed on the particles than before, and the intermetallic compounds of the core PdFe were dissolved/leached. The core protected by the Pt shell was well preserved, while that which was not covered underwent severe leaching.

Thus, we can conclude that ordered intermetallic compounds with a core–shell structure not only effectively improve ORR activity due to the ligand effect and strain effect, but also availably prevent the dissolution/leaching of internal intermetallic compounds with a Pt shell to improve the stability of the catalyst [121,127,129]. What is noteworthy is that such core–shell Pt-based intermetallic catalysts also have remarkable performance in improving ORR performance on MEA. Cheng et al. [123] used ultrafine Pt nanocrystals as the shell to prepare a core–shell structure Pt₁Co₁ intermetallic compound (Pt₁Co₁-IMC@Pt) with a size of less than 6 nm. Thanks to the core-shell structure and the highly ordered arrangement of Pt-Co atoms, the d-band center on the Pt surface was significantly reduced, and the oxidation resistance of the Pt/C sites was enhanced. As a result, after the RDE test, the catalyst showed excellent electrochemical performance with the mass activity being 0.53 A mg_Pt⁻¹, and it only decreased by 23.4% after 30,000 ADT tests. And after 30,000 cycles of the AST test, the mass activity of the crystal was only reduced by 24.8%, which was similar to the ADT result measured using the RDE. Additionally, when using this catalyst in the PEMFC, the cell had a high power density of 2.30/1.23 W cm⁻² under H₂-air/O₂ conditions at 80 °C. Furthermore, [email protected] V was 0.46 A mg_Pt⁻¹ in the MEA test, which is close to the intrinsic value of the RDE measurement and exceeds the goal set by the DOE (0.44 A mg_Pt⁻¹). Similarly, Pan et al. [130] designed an intermetallic catalyst (L1₀-CoPt@Pt) composed of L1₀-CoPt as the core and Pt shell. Due to the large ECSA caused by the compressive strain and small size of the Pt shell surface, it exhibited excellent activity in MEA, and the mass activity was 0.6 A mg_Pt⁻¹, which is significantly more than the target of 0.44 A mg_Pt⁻¹ set by the DOE. Meanwhile, when it was used in the H₂/air fuel cell, the power density increased to 0.986 W cm⁻² at 0.6 V/150 kPa_abs. In addition, it had noticeable stability (MA loss was only 40% after 30 k AST) because the stable core–shell structure can significantly increase the diffusion barrier of Co and effectually prevent its leaching. The preparation processes of core-shell structure materials are complicated, and the ECSA of most ordered intermetallic compounds with a core–shell structure is much smaller than that of a commercial Pt/C catalyst.

Figure 3. (a) A schematic illustration of the random alloy and intermetallic nanocrystals. Reprinted with permission from Ref. [111]. Copyright 2020, the American Chemical Society. (b) A schematic diagram of PtFe disordered catalysts and Pt skin-ordered PtFe catalysts. Reprinted with permission from Ref. [126]. Copyright 2015, the Royal Society of Chemistry.

Figure 3. (a) A schematic illustration of the random alloy and intermetallic nanocrystals. Reprinted with permission from Ref. [111]. Copyright 2020, the American Chemical Society. (b) A schematic diagram of PtFe disordered catalysts and Pt skin-ordered PtFe catalysts. Reprinted with permission from Ref. [126]. Copyright 2015, the Royal Society of Chemistry.

Currently, researchers are able to enhance the ORR performance of Pt-based intermetallic catalysts by adjusting the crystal structure [52,116]. Nevertheless, with more in-depth research, it is recognized that the intrinsic activity of the intermetallic catalyst is more affected by the ordering degree [51,54]. To be more specific, increasing the degree of order can dramatically enhance the interaction between Pt and transition metals and optimize the electronic structure of the Pt surface to make the d-band center in a more suitable position, which is beneficial to improve the ORR performance [112,131,132]. In order to further clarify the relationship between the ordering degree and the catalytic activity of the intermetallic catalyst, Song et al. [51] prepared a series of PtFe intermetallic compounds with different ordering degrees and similar particle sizes (Figure 4a). Among them, the PtFe-H/Pt catalyst with the highest ordering degree manifested the highest mass activity (2.212 A mg_Pt⁻¹) (Figure 4b,c). Therefore, it can be determined that the ordering degree is positively correlated with ORR performance (Figure 4d). When applied to H₂/O₂ fuel cells, it also exhibited high mass activity (0.92 A mg_Pt⁻¹ at V_iR-corrected), which was more than twice as high as the DOE standard (0.44 A mg_Pt⁻¹) (Figure 4e). And after 30 k ADT cycles, MA only decreased by 24%, which met the standard set by the DOE (the loss of MA less than 40%) (Figure 4f). The excellent performance can be attributed to the highly ordered structure and compressive strain effect of intermetallic compounds. Lei et al. [131] believed that the high ORR performance of ordered intermetallic compounds is due to their low lattice constants, which cause surface compressive strain, thereby reducing oxygen adsorption energy. Increasing the ordering degree can further reduce the lattice constant and greatly improve the activity. Moreover, the d electron orbits of the Pt atom and transition metal M in the ordered structure are highly overlapped, which can availably inhibit the leaching of the transition metal and maintain the stable crystal configuration. But recent studies have shown that the durability of catalytic nanoparticles is more affected by particle size and distribution. Sandbeck et al. [133] studied the effect of particle size on the solubility of platinum by the AST test. The dissolution trend indicated that the smaller size promoted the formation of a PtO_x passivation layer, thereby inhibiting the dissolution of nanoparticles and showing higher stability. Therefore, maintaining a high ordering degree and uniform distribution under a small size is an important way to improve the intrinsic ORR performance of intermetallic catalysts, whereas during the high-temperature annealing process of forming an ordered structure, it is inevitable to cause Ostwald ripening, which leads to the serious aggregation and sintering of intermetallic compounds as well as changes the size and distribution of catalytic nanoparticles. To attain the intermetallic catalyst’s maximum ordering degree and lowest particle size, Liang et al. [54] studied the structural evolution of PtFe, PtCo, and PtNi intermetallic compounds during the synthesis process to determine the alloying and ordering process. By controlling the annealing process to achieve the separated high-temperature alloy and low-temperature ordering stage, the obtained PeFe-ordered intermetallic compound took both the small size and high order degree into account, showing a high mass activity of 2.61 A mg_Pt⁻¹ and only a decrease of 30% after 30 k cycles. It was further tested under MEA operating conditions, and it also exhibited a high mass activity of 0.96 A mg_Pt⁻¹, which is much higher than that of commercial Pt/C catalysts (0.20 A mg_Pt⁻¹), and the MA remained at 84% after 30 k ADT cycles. This can be attributed to the high ordering degree and ECSA of PtFe-ordered intermetallic catalysts. Furthermore, the addition of molecular additives containing heteroatoms can also solve the sintering problem caused by high-temperature annealing [122,134,135]. For example, Song et al. [53] used a small molecule-assisted impregnation approach to coordinate platinum with heteroatoms (O, N, or S). Since the formed heteroatom-doped graphene layer can encapsulate and anchor metal nanoparticles, it was able to solve the sintering problem during the high-temperature annealing process. The obtained intermetallic compounds ensured a small size in the case of a high order and displayed excellent performance in both the RDE and MEA tests. In particular, when PtCo intermetallic compounds were used as PEMFC cathode catalysts, they exhibited a high MA (1.08 A mg_Pt⁻¹), exceeding the U.S. DOE’s 2025 target (0.44 A mg_Pt⁻¹), and the MA only decreased by 25% at 30 k cycles. Moreover, the synergistic metal–support interaction and the space confinement effect are also common strategies to avoid the agglomeration of metal nanoparticles. Zhang et al. [136] reported a Pt₃M/rGO-HF catalyst prepared by the hydrogel freeze-drying strategy. Benefiting from the confinement effect of three-dimensional porous material (rGo), the Pt₃M/rGO- HF catalyst had an ultra-small size (less than 3 nm) and a uniform particle distribution. In the ORR test, it exhibited high mass activity (0.233 A mg_Pt⁻¹), which was much larger than that of the disordered D-Pt₃Mn/rGO-HF catalyst (0.218 A mg_Pt⁻¹) and commercial Pt/C catalyst (1.70 mA mg_Pt⁻¹), and the half-wave potential only decreased by 15 mV after 5 k cycles, confirming its high stability.

Although people have conducted extensive research in recent years, there are still some problems and vacancies that need to be further improved. Thus, the following could be the future directions for study and development [44,137]:

• During synthesis and preparation, there may be some problems in the high-temperature annealing process, such as the aggregation of nanoparticles, the difficulty in forming a core–shell structure for some transition metals (such as Fe), and trouble in controlling the shape of intermetallic compound nanocrystals during synthesis. Researchers need to explore the formation mechanism of nanocrystals from the perspective of reaction kinetics and thermodynamics, understand the relationship between surface structure and catalytic performance at the atomic scale, and develop more low-temperature annealing methods on this basis. In the future, adding low-melting metals may be the direction for structural ordering under milder conditions.
• Theoretical calculations can be used to further guide the determination of the morphology/structure of ordered intermetallic compounds and the study of the kinetic process of the catalyst surface.
• The reaction process can be monitored using advanced characterization techniques, such as in situ infrared spectroscopy (FTIRS), transmission electron microscopy (TEM), X-ray diffraction (XRD), and synchrotron radiation (XAFS) in in situ characterization techniques so as to further and more thoroughly study the reaction mechanism and deactivation mechanism.

4. Pt-Based High-Entropy Alloy Catalysts

High-entropy alloys (HEAs) have attracted much attention in the field of catalysis since they were first reported in 2004 [56]. The definitions of high-entropy alloy catalysts are still controversial, which are mainly explained from the two aspects of composition and entropy. In terms of composition, high-entropy alloy catalysts are composed of five or more elements, and the atom proportion of each element is about 5% to 35% [56,57,138]. From the perspective of entropy, the mixed configuration entropy of alloy catalysts can be calculated using the following formula:

$$S=-R\sum x_i\ln x_i$$

where $R$ is the molar gas constant; $x_i$ represents the molar fraction of each component.

In general, alloy catalysts composed of five or more elements can be considered as high-entropy alloy catalysts when the mixed molar entropy $S\ge 1.5R$ [56,57,139,140,141]. Normally, high-entropy alloy catalysts can meet the above two definitions.

In contrast to conventional alloys made of two or three metals, high-entropy alloy materials show more excellent properties [142,143,144], such as high catalytic activity, strong corrosion resistance, and high hardness and strength. This is because high-entropy alloy materials are affected by the high-entropy effect, the lattice distortion effect, the sluggish diffusion effect, and the cocktail effect [58]. Specifically, when there are multiple elements mixed, the configurational entropy of the system increases significantly, and the high-entropy effect plays a main role [56,57,143]. In high-temperature synthesis, compared with the formation of intermetallic compounds, it is easier for the system to form a simple single-phase solid solution, such as an FCC phase, BCC phase, and HCP phase, which makes HEAs have excellent stability [57,138,140,145,146]. Secondly, due to the substantial variation in the atomic radius of HEAs and the random distribution in the lattice, the high-entropy alloy catalyst has the serious lattice distortion effect [58,147,148]. The atoms deviating from the equilibrium position in the distorted lattice will increase the dislocation resistance. This improves the hardness of the materials while facilitating the formation of defects and strains, which shows higher catalytic activity [142,149,150]. In addition, the serious lattice distortion effect causes the atom to be in a non-equilibrium state with high energy, and the diffusion energy barrier is significantly increased, which hinders the surface diffusion of the atom and makes it have the sluggish diffusion effect [141,151,152,153,154]. Therefore, HEAs can still maintain high stability in harsh environments such as strong acid corrosion [155]. Additionally, the interactions and synergies between different elements in high-entropy alloy materials can produce various unexpected and peculiar effects, which was defined as the cocktail effect by Professor S. Ranganathan [138,156]. High-entropy alloy materials exhibit high strength and corrosion resistance, which are related to the cocktail effect [55]. However, the mechanism of this effect still needs to be further elucidated by theory and experiment. This also shows that HEAs have great development potential in electrocatalytic applications.

At present, there have been many reports on the application of HEAs in electrocatalytic reactions. Here, we mainly review its application in ORR. Researchers mainly improve the performance of the catalyst based on two aspects: structure and composition [149,157]. In more detail, due to the complex microstructure of HEAs, the crystal structure plays a decisive role in its catalytic performance [141,157,158]. Starting from the structure–activity relationship of HEAs, researchers have increased the reaction surface area and exposed surface of the catalyst by controlling the geometric configuration of the crystal, such as the convex cube structure [157], hollow structure [159], nanowires [160], nanosheets [161], and so on. This can maximize the adsorption sites of the reaction intermediates and reactants so that it has a lower catalytic reaction energy barrier. In terms of structural control, Chen et al. [157] reported a Pt₃₄Fe₅Ni₂₀Cu₃₁Mo₉Ru HEA catalyst. Due to the doping of Ru, the catalyst had a unique convex cubic shape and a high-index exposed surface. Compared with the (111) plane of the pure Pt catalyst, the catalytic performance was significantly improved, with a half-wave potential of 0.87 V and a limiting current density of 5.6 mA cm⁻². The lower Tafel slope (69 mV dec⁻¹) further confirmed its excellent catalytic activity. After 40 h under a constant current electrolysis test, the current density remained at 89%, indicating its excellent stability. The DFT calculations show that the synergistic effect and strain effect between each element, convex cube structure, and Ru content are beneficial to the interaction of the catalyst and the reaction intermediate, thereby improving the activity and stability of the catalyst. However, the investigation of the structure–activity relationship of HEAs is hampered by the complex model. There are still few studies on the different crystal shapes of HEAs and their relationship with properties.

Furthermore, utilizing the surface effect of the catalyst, when the particle diameter is reduced to the nanometer level, the increase in the specific surface area can produce more surface-active exposure sites and improve the performance of the catalyst. Chen et al. [162] synthesized PtFeCoNiCu high-entropy alloy nanoparticles with a uniform element distribution and small size (about 5 nm) via an impregnation reduction method. Due to the existence of multiple active sites and optimized electronic structure on the surface, the catalyst exhibited excellent electrocatalytic activity (1.738 A mg_Pt⁻¹) and stability (MA remains 92% after 10 k cycles test). Additionally, when PFCNC-HEA catalysts are used as PEMFC cathode catalysts, the maximum power density of the cell could reach 1.38 W cm⁻². On this basis, the author of [59] further minimized the nanoparticle size and reported an ultra-small (about 2 nm) PtFeCoNiCuZn-ordered high-entropy intermetallic (PFCNCZ-HEI) (Figure 5a,b). Benefitting from a small particle size and metal surface electronic structure optimization, catalysts with multiple active sites showed a high mass activity (2.403 A mg_Pt⁻¹) at 0.9 V in the RDE test, which is more than 19 times that of the commercial Pt/C catalyst (Figure 5c). Moreover, the high-entropy effect, stable structure, and spatial restriction effect of the Zn-DPCN support on the nanoparticles enabled PFCNCZ-HEI catalysts to maintain 94.1% mass activity after 10 k cycles, suggesting long-term durability (Figure 5d). When they were applied to the cathode catalysts of PEMFCs, the power density of the fuel cell can reach 1.4 W cm⁻² (Figure 5e) and exhibit long-term stability (Figure 5f). In particular, the mass activity of PFCNCZ-HEI catalysts at 0.9 V (1.1 A mg_Pt⁻¹) exceeded the U.S. Department of Energy’s target (0.44 A mg_Pt⁻¹). This also provides a new idea for the improvement in catalyst performance through size control.

Compared with Pt alloy catalysts with a large immiscible gap, the high-entropy effect of HEAs significantly reduces the immiscible gap between metals [163,164]. Therefore, researchers use the surface strain effect and coordination effect between polymetallic atoms to change the local coordination environment of the atoms by precisely adjusting the element type and composition of the HEA catalyst [149,165]. Thus, the number and intrinsic activity of reactive sites are increased. Bathelor et al. [166] used HEAs with an almost continuous adsorption energy; the adsorption energy of the reaction intermediates *OH and *O at the random binding sites on the surface of the high-entropy IrPdPtRhRu alloy catalyst was obtained by DFT calculations; and the adsorption energy database was constructed. On this basis, an HEA catalyst was used as a design platform to construct the binding site near the peak of the volcano map by precisely adjusting the catalyst composition. Compared with Pt/C catalyst, the adsorption energy of the surface-active sites of the optimized Ir_10.2Pd_32.0Pt_9.30Rh_19.6Ru_28.9 HEA catalyst was closer to the peak of the volcanic curve. Meanwhile, the activity was doubled, and the overpotential was about 40 mV lower than that of the (111) surface of the pure Pt catalyst, which indicated that it had better electrocatalytic performance. Liu et al. [149] reported a class of five- to seven-phase HEA mesoporous nanotubes (HEA mNTs). The authors incorporated different transition metals M (Mn, Fe, Co, or Ni) into PtPdRuIrMCu mNTs. Through DFT calculations and experiments, it was concluded that the iron-regulated PtPdRuIrFeCu mNT catalyst had great catalytic activity, and the mass activity reached 1.94 A mg_Pt⁻¹, which is 7.46 times higher than that of the commercial Pt/C catalyst. This is because the HEA mNTs’ structure contained fewer atomic layers, which maximized the exposure of atomic active sites, and many high-catalytic-activity unsaturated sites were generated due to lattice distortion and abundant defect sites. Li et al. [165] found a np-AlCuNiPtMn catalyst with high ORR activity through elemental composition control and screening. They uniformly dispersed five immiscible metals into the nanosolid phase using a top-down dealloying method. A series of np-HEA with a low Pt content (20–30%) was obtained through the pre-determination of Al, Cu, Ni, and Pt contents and by alternating the addition of the fifth element (Pd, V, Co, and Mn). Among them, the np-AlCuNiPtMn catalyst exhibited excellent electrocatalytic activity and stability. The mass activity in acidic medium was 3.5 A mg_Pt⁻¹, which is 16 times that of the Pt/C catalyst (0.22 A mg_Pt⁻¹), and the highest half-wave potential was 0.945 V. This further indicates that there is still room for HEA catalyst design—a novel approach to HEA synthesis will be recommended.

Adjusting the structure so that it becomes ordered is an innovative method to improve the ORR performance of high-entropy alloy catalysts. The ordered high-entropy alloy catalyst can significantly reduce the uncontrollability caused by the random distribution of HEA active sites for catalyst structure design and component adjustment [55,158]. In theory, the synthesized multi-component intermetallic catalyst has the advantages of both a high-entropy alloy catalyst and intermetallic catalyst. Wang et al. [167] obtained a class of Pt₄FeCoCuNi high-entropy intermetallic catalysts with an adjustable ordering degree by controlling the annealing temperature and time. The test results show that the ordered Pt₄FeCoCuNi catalysts had excellent catalytic activity with a mass activity of 3.78 A mg_Pt⁻¹, which was better than those of the partially ordered sample (1.84 A mg_Pt⁻¹) and disordered sample (0.86 A mg_Pt⁻¹). The result reveals that the catalyst activity increased with the increase in the ordering degree. This can be attributed to the fact that the ordered high-entropy intermetallic catalyst has a controllable degree of order to optimize the electronic structure and surface structure of the catalyst, resulting in the weakening of the binding strength of the reactants and the downward shift of the d-band center. In the durability test, the mass activity of the ordered sample only decreased by 26% after 30 k cycles, and the partial ordered and disordered samples decreased by 44% and 62%, respectively, signifying that the ordered multi-element catalyst had higher stability. This was the outcome of combining the high-entropy effect with the highly ordered crystal structure. This relationship between ordered structure and performance provides a new approach for the design of multi-component metal catalysts. However, the high-temperature annealing method can easily lead to the rapid growth and aggregation of catalytic nanoparticles. Loading them in mesoporous voids can effectively inhibit the sintering of the catalyst. Luo et al. [161] successfully constructed ordered HEA NPs on a novel 2D nitrogen-doped mesoporous carbon nanosheet (OHEA-mNC) by combining ligand-assisted interface assembly with NH₃ annealing. The test results show that, compared with the disordered HEA catalyst and commercial Pt/C catalyst, the OHEA-mNC catalyst exhibited higher catalytic activity (E_1/2 was 0.9 eV) and stability (E_1/2 decayed by 10 mV after 10,000-cycle test). The reason was that the unique-ordered intermetallic phase structure of the catalyst induced the d-band center to move downward to the Fermi level, thereby changing the adsorption mode of the reactants or intermediates. Moreover, the metal elements were uniformly dispersed, providing a surface structure that was easier to react while maintaining structural integrity during the catalytic process. Feng et al. [60] reported a structurally ordered PtIrFeCoCu HEI catalytic nanoparticle (PIFCC-HEI NP). In the ORR test, the PIFCC-HEI/C catalyst exhibited ultra-high mass activity (7.14 A mg_{noble metal}⁻¹), which was 20.4 times and 2.4 times those of the commercial Pt/C catalyst (0.35 A mg⁻¹) and PIFCC-HEA/C catalyst (3.03 A mg_{noble metal}⁻¹), respectively. Additionally, E_1/2 only decreased by 9 mV in 60 k ADT, which meant that the stability was significantly higher than that of the commercial Pt/C catalyst (decreased by 49 mV). The PEMFC obtained by using the PIFCC-HEI NP catalyst as a cathode ORR catalyst also exhibited top performance, and the current density (4.25 A cm⁻²) at 0.4 V was much higher than that of the commercial Pt/C catalyst (3.67 A cm⁻²). More importantly, the maximum peak power density could reach 1.73 W cm⁻², and the operating voltage did not decay after 80 h of operation. These excellent properties were mainly due to the structurally ordered multi-element HEI, which converted the low active surface (001) into the ultra-high active surface, which could minimize the active barrier of ORR while lowering the d-band center to the best position and optimizing the surface electronic structure of the catalyst. This work also shows the great development potential of ordered high-entropy alloy catalysts for practical fuel cells. However, studies have rarely been conducted on the reaction mechanism of high-entropy alloy catalysts from the disordered to ordered structure.

Although HEA catalysts have shown promise, the current research is still in its infancy, and there are still many problems to solve in terms of reaction mechanism, catalyst design, and synthesis strategy:

• The reaction mechanism and pathway of the catalyst have not been fully clarified; for example, the unique reaction mechanism of the cocktail effect and the connection between lattice distortion and activation barrier, etc., still need to be further studied. In addition, the reaction mechanism of a high-entropy alloy catalyst from a disordered to ordered structure has still not been explored. Current research often uses simple models to illustrate complex processes. Therefore, more advanced in situ/operational characterization methods, such as operando TEM and SEM techniques, are needed to understand the structure–activity relationship of HEA and to guide the design of catalysts [148,168].
• The rational design of high-entropy alloy catalysts is also a challenge. The active centers of Pt-based HEAs are composed of Pt and its surrounding atoms, but there are many possible atomic arrangements of Pt atoms. It is challenging to identify the active center of HEAs and calculate the ideal composition ratio due to the various coordination environment that each Pt atom has. Therefore, the design and preparation of a catalyst structure and composition require more advanced computer technology to assist in judging the reaction center and predicting the results. In the future, with the development of advanced methods such as machine learning and statistical analysis, it is expected that we can accurately and efficiently predict and design the composition and performance of high-entropy alloy catalysts [141,169,170].
• Common synthesis methods and routes have low efficiency. However, it is well known that the development of efficient synthesis methods and routes to improve the yield is the key to HEA catalysts from experimental research to large-scale industrial production. At present, the synthesis reaction of an HEA catalyst usually needs to be carried out under a high temperature, high pressure, and inert gas, which requires high experimental equipment and conditions. For example, the fast-moving bed pyrolysis method requires rapid heating and cooling in a specific instrument to uniformly disperse the catalyst at high temperatures with very small nanoparticles. Although electrochemical deposition synthesis can solve this problem well and control the reaction by adjusting the applied voltage at room temperature, the obtained materials are unevenly distributed and cannot be mass-produced. Therefore, efficient and controllable multi-component catalyst synthesis methods still need to be developed [171,172].

5. Coupled Low-Pt and PGM-Free Catalysts

Platinum group metal (PGM) catalysts have achieved high ORR performance at a low platinum loading, also known as low-Pt catalysts. However, reducing the platinum content will inevitably lead to the loss of catalytic active sites and hinder local oxygen transfer, resulting in limited catalyst performance [173,174,175,176]. Researchers found that the PGM-free catalyst represented by metal–nitrogen–carbon (M-N-C, M: Fe, Co, Mn, etc.) showed low stability and slow ORR kinetics in PEMFCs [62,177,178,179]. However, it is often manufactured using a template made of porous polymer or metal–organic frameworks, which has an easily adjustable pore structure and a large specific surface area. It can solve the problem of the slow oxygen mass transfer rate of low-Pt catalysts, providing sufficient channels for oxygen transport and supplement catalytic active sites [62,180,181,182]. In order to accurately grasp the interaction between low-Pt catalysts and PGM-free catalysts, Chong et al. [61] quantitatively detected the PtCo@Co-N_xy catalyst and found that the ORR activity of the PtCo@Co–N_xy catalyst was significantly higher than that achieved with the simple addition of PtCo catalyst and Co–N_xy catalyst, which proved that low–Pt catalysts and PGM–free catalysts were coupled for ORR catalysis. The authors also attributed the high stability of the catalyst (the MA remained at 64% after the 30 k cycle test) to the coupling effect between low-Pt catalysts and PGM-free catalysts. Similarly, Xia et al. [46] reported the PtCo alloy catalyst encapsulated by Co–nitrogen–graphene. The synergistic effect between the PtCo alloy and Co-N₄ in the coupling catalyst significantly increased the catalytic active sites, and the strong interaction between them effectively reduced the adsorption energy of oxygen-containing intermediates, thereby improving the stability of the catalyst. Therefore, coupling low-Pt catalysts with PGM-free catalysts to form mixed catalysts is a promising strategy to further improve ORR performance.

The coupling effect between low-Pt catalysts and PGM-free catalysts is beneficial to improve the selectivity of ORR [173,183,184]. Currently, researchers generally believe that PGM-free catalysts tend to generate hydrogen peroxide in the two-electron (2e⁻) ORR pathway when they are independently catalyzed in PEMFCs [185,186,187]. However, it has been reported that coupled low-Pt and PGM-free catalysts exhibit high selectivity for the four-electron (4e⁻) reaction pathway [187,188,189,190,191]. Xiao et al. [192] evaluated the Pt/Fe-N-C catalyst in a rotating ring–disk electrode (RRDE). By comparing the ring current of the Pt/Fe-N-C catalyst with that of the Fe-N-C catalyst, it was proven that the Pt/Fe-N-C catalyst had a lower yield of hydrogen peroxide in the catalytic process and showed higher 4e⁻ reaction selectivity. However, the reaction mechanism was not clarified in this paper. In this regard, Lai et al. [193] believe that the key to enhance 4e⁻ selectivity and the ORR rate lies in the rapid migration of H₂O₂ generated at the Co-N₄ site to L1₀-PtCo and further decomposition into H₂O while reducing the ORR energy barrier. Likewise, Zhao et al. [194] concluded through DFT calculations that the 10% Pt/C-N-C catalyst showed a low H₂O₂ yield because H₂O₂ overflowed from the Co-N_x site to the Pt(111) surface. In addition, the synergistic catalysis of multiple active sites in coupled low-Pt and PGM-free catalysts is also beneficial to reduce the adsorption energy of oxygen and promote the efficient 4e⁻ ORR pathway. For example, Yang et al. [195] manifested that the synergistic effect between Pt/C and Fe/N/C promoted the process of oxygen adsorption and the desorption of oxygen-containing intermediates. The test results show that the half-wave potential of the Pt/C-Fe/N/C catalyst was higher than those of the Fe/N/C and Pt/C catalysts. Additionally, the MEA performance of the Pt/C-Fe/N/C catalyst (0.22 A mg_Pt⁻¹) was much higher than that of the Pt/C catalyst (0.02 A mg_Pt⁻¹) in the fuel cell test. This also indicates that the coupled catalyst can perform ORR more efficiently.

Additionally, in previous reports, researchers found that M-N-C (PGM-free catalysts) can be used to optimize the surface structure of Pt-catalyzed nanoparticles to produce coupled catalysts with better ORR activity [108,196,197,198]. Zhao et al. [194] reported the mixed catalysts coupled by Pt nanoparticles and Co-N_x catalysts. Among them, the 10% Pt/Co-N-C catalyst with the best performance showed high mass activity (0.223 A mg_Pt⁻¹), which was 2.8 times that of the 20% Pt/C catalyst. The improved intrinsic activity of the catalyst could be attributed to the electronic synergistic effect between Co-N_x and Pt single atoms, which optimized the electronic structure of the Pt surface and significantly reduced the adsorption energy of oxygen-containing intermediates. Furthermore, the catalyst exhibited remarkable stability, with nearly no reduction in half-wave potential during the ADT test. The reason is that the interaction between Pt nanoparticles and the Co-N-C catalyst can anchor Pt nanoparticles to prevent their growth and agglomeration. Similarly, Xiao et al. [199] reported that the Pt/Fe-N-C catalyst had comparable catalytic activity and excellent durability (the ECSA retention rate was as high as 99% after the 10 k cycle test). This is because Fe-N-C can provide stable support for Pt atoms and reduce the formation of oxygen-containing intermediates by optimizing their surface electronic structure. Nevertheless, due to the low optimization degree of M-N-C for the surface electronic structure of Pt nanoparticles and the easy dissolution of Pt NPs under the MEA operation, the coupling catalysts formed by Pt nanoparticles and M-N-C still cannot meet the requirements of PEMFCs for catalyst activity and stability. In this regard, researchers performed alloying operations to achieve the effective regulation of the surface electronic structure of Pt nanoparticles and enhanced coupling effects [189,190,191,200,201]. Xia et al. [63] formed a Co-carbon integrated catalyst (PtCo@CoNC/NTG) (Figure 6a) by integrating a PtCo alloy on a cobalt–nitrogen–carbon nanoscale (Figure 6b) through multi-scale design engineering. Thanks to the synergistic catalysis of the PtCo alloy and multiple active sites of Co-N-C, it had excellent catalytic activity, and the mass activity (1.52 A mg_Pt⁻¹) was 11.7 times higher than that of the commercial Pt/C catalyst (0.13 A mg_Pt⁻¹) (Figure 6d). In addition, coupling between the PtCo alloy catalyst and Co-N-C catalyst formed a stable spatial structure (Figure 6c) and maintained 98.7% stability after the 30 k potential cycle test, showing excellent durability (Figure 6e). In order to study its MEA performance, the catalyst was used in a hydrogen–air fuel cell to measure a current density of 1.50 A cm⁻² at 0.6 V and a power density of 980 mW cm⁻² (Figure 6f). The fuel cell was continuously stably operated for 24 h (Figure 6g). It is worth noting that the polarization current (0.308 A cm⁻²) of the PtCo@CoNC catalyst at mV_{iR-uncorrected} exceeded the target set by the DOE. This can be attributed to the unique 3D network structure in the PtCo@CoNC/NTG catalyst, which has a rich pore structure to provide more material transport channels for the reaction, significantly accelerating the oxygen diffusion rate and thus allowing for efficient catalysis in hydrogen–air fuel cells.

In order to further improve the ORR performance, researchers used ordered intermetallic catalysts with a better surface electronic structure and more stable structure compared to low-Pt catalysts [189,202]. Qiao et al. [108] loaded Pt nanoparticles and ordered L1₂ Pt₃Co intermetallic nanoparticles on FeN₄-C, respectively, to obtain the coupled Pt/FeN₄-C and Pt₃Co/FeN₄-C catalysts. In both the RDE and MEA tests, the coupled Pt₃Co/FeN₄-C catalysts showed better catalytic activity and stability than the Pt/FeN₄-C catalysts. Under MEA conditions, the Pt₃Co/FeN₄-C catalyst had high mass activity (0.72 A mg_Pt⁻¹), which was much higher than the U.S. DOE’s standard (0.440 A mg_Pt⁻¹), and it had a high power density (824 mW cm⁻² at 0.67 V). This was because the porous structure of FeN₄-C can control the uniform dispersion of the intermetallic compounds at a small size (about 4 nm), which can strengthen the synergistic effect between L1₂Pt₃Co and FeN₄-C. In addition, the coupled Pt₃Co/FeN₄-C catalyst demonstrated good stability after AST testing (21 mV voltage loss at 0.8 A cm⁻²), exceeding the U.S. DOE’s stability target (30 mV loss). This can be attributed to the enhanced metal–support interactions and the ordered structure of intermetallic compounds. However, it is still difficult to create high-performing coupled low-Pt and PGM-free catalysts with small sizes because the size change in the nanoparticles during high-temperature annealing to form the ordered structure weakens the interaction between the low-Pt and PGM-free catalysts [108,180,202,203]. To this end, Guo et al. [204] adopted a microwave-assisted heating program to first load Pt nanoparticles on Co-N-C and then disperse Co atoms into the lattice of Pt under a heat treatment in order to form ultra-small-sized intermetallic nanoparticles. Benefiting from the suitable d-band center and enhanced metal-support interaction, the PtCo/Co-N-C catalyst possessed high catalytic activity (0.700 A mg_Pt⁻¹@0.9 V) and durability (only 16 mV loss in E_1/2 after the 30 k cycle test). It was also used as a cathode catalyst for fuel cells, exhibiting high power density (0.70 W cm⁻²) at a low loading of 0.05 mg cm⁻², which is 0.14 W cm⁻² higher than that of the Pt/C catalyst under the same conditions. This also provides a new method for the realization of the coupling catalyst with a low platinum loading at a small particle size.

Although coupled low-Pt and PGM-free catalysts can provide a new path to further improve catalyst performance, research on these catalysts is still in the initial stage, and there are many problems, such as the following:

• The complex structure makes it difficult to detect the number of active sites accurately and quantitatively. At present, there is a deviation in the number of active sites of PGM-free catalysts detected by Mössbauer spectroscopy, CO chemisorption, and other methods. Fourier transform alternating current voltammetry was reported by Rifael Z et al. [205]; it may be a promising method for accurately measuring the density of PGM-free active sites in the future. Additionally, the number of active sites can be determined by advanced artificial intelligence and computers to guide catalyst design [206,207,208,209].
• The coupling mechanism of coupled low-Pt and PGM-free catalysts has not been fully elucidated, especially the mechanism conducive to the selectivity of the four-electron ORR pathway. At present, these properties are mostly verified from experiments, and a comprehensive understanding of the coupling effect still needs to be obtained. In the future, advanced in situ characterization and operational characterization techniques can be used to accurately grasp the electrochemical reaction process of coupled low-Pt and PGM-free catalysts and construct the relationship between performance and structure [62,206].

6. Single-Atom Pt Catalysts

Compared with traditional catalysts, single-atom Pt catalysts are the new type of Pt-based catalytic material that disperses metals in the form of single atoms on the support. Pt single atoms show broad development prospects in electrocatalysis. As the particle sizes are greatly reduced, the energy level structures and electronic structures change, and the atomic utilization rates are greatly improved [64,210]; theoretically, they can be 100%. Moreover, the ample unsaturated coordination sites and unique electronic structures on the single atoms make them have higher catalytic activity and selectivity. However, similar to PGM-free catalysts, Pt single atoms have high selectivity for the two-electron (2e⁻) ORR pathway. It is well known that in the ORR, oxygen can take a four-electron path to generate water or a two-electron path to generate hydrogen peroxide, and the direct four-electron ORR enables energy to be converted and used in a more efficient manner. The previous paper also mentioned that only when oxygen-containing intermediates are properly adsorbed and desorbed on the surface of the catalyst can the reaction take a four-electron path. However, it was found that when the particle size of the Pt catalyst was reduced to the atomic level, the selectivity of the Pt single atom for 2e⁻ ORR pathway increased greatly due to the slow oxygen adsorption and dissociation on the surface of the Pt single atom, and it could not independently promote the 4e⁻ ORR pathway. This seriously restricts the development of the single-atom Pt catalyst as an ORR catalyst. [26,211,212,213].

Despite the belief that the Pt single-atom catalyst is unable to perform 4e-ORR on its own, it still has a certain development prospect in the field of PEMFC cathode catalysts. The abundant active sites on the surface of Pt single atoms and the synergistic effect of metal–metal can help the traditional Pt/C catalyst that tends to generate a 4e⁻ pathway to achieve high ORR performance [25,214] Wang et al. [66] conducted a study based on the fact that Pt single atoms and nanoparticles coexist inevitably in Pt/C catalysts. After conducting theoretical calculations and experiments on the single-atom Pt catalyst (Pt_SA/C), the nanocluster catalyst (Pt_NC/C), and the combination of the two catalysts (Pt_SA-NCX/C), it was found that the synergistic effect between the single atom and nanocluster can optimize the adsorption and desorption process of an oxygen-containing intermediate (*OOH), and it significantly reduced the reaction energy barrier. Similarly, Dong et al. [215] prepared a Pt/C catalyst with the best hybrid effect by optimizing the mass ratio between Pt nanoparticles and single atoms in which the content of Pt was 8 wt% (5 wt% Pt_SAs, 3 wt% Pt NPs). The test data suggest that doped Pt nanoparticles can inhibit the formation of H₂O₂ on the surface of Pt single atoms, and the synergistic effect between the two can promote the four-electron ORR and exhibit high catalytic activity, which is almost equivalent to the catalytic activity of the Pt/C catalyst with a 20 wt% Pt loading. This work also provides a new idea for Pt single atoms to improve the activity of traditional Pt/C catalysts.

Likewise, transition metal single atoms (Fe, Co, Cu, etc.) are uniformly dispersed on the surface of the Pt catalyst to form the metal-supported single-atom catalyst, which can also accelerate the ORR process of the Pt-based catalyst [25]. Zhang et al. [65] reported a Co single atom-modified Pt catalyst (Co SA-modified Pt NPs). The Pt-Co bond formed on the Pt surface changed the electronic structure and weakened the adsorption energy of O * and OH *. These can significantly improve the ORR activity (mass activity of 0.49 A mg_Pt⁻¹) and stability (MA decreased by 20.8% after 10 k cycles). Researchers have found that the combination of single-atom metals with intermetallic catalysts can further optimize ORR activity. Zeng et al. [216] synthesized a low-platinum catalyst by loading Pt nanoparticles on the single Mn site-rich carbon (Figure 7a). The obtained Pt@Mn_SA catalyst (Figure 7b) has excellent ORR activity and durability (Figure 7d,e). Under MEA conditions, it achieved a high mass activity of 0.63 A mg_Pt⁻¹ (0.9 V_iR-free) and 1.05 A cm⁻² (0.7 V, H₂–air fuel cell), with a loss of 22% and 9% after 30 k cycles of the AST, respectively. In addition, a compact Pt skin coating with 2–3 atomic layers (arrows in Figure 7c) on the ordered L1₂₃Co@Mn_SA intermetallic catalyst (Figure 7c) obtained by loading uniformly dispersed Pt and ordered L1₂₃Co nanoparticles on the Mn_SA support can further improve the ORR performance (Figure 7d,e). The initial current density was 1.75 A cm⁻² at 0.7 V in the HDV MEA (Figure 7f), and the MA only lost 18% at 90 k cycles of AST (Figure 7g), indicating that it had great potential to meet the DOE target of HDV. According to the experimental results and DFT calculations, the MnN₄ single atom maximized the interaction between Pt and the support to reduce the agglomeration of Pt and significantly reduced the adsorption energy of oxygen-containing intermediates on the surface of Pt, which further improved the ORR intrinsic activity. Likewise, Wang et al. [217] developed an atomically dispersed low-platinum catalyst (5.5 wt% Pt loading) with FeN₄-rich sites (Pt@Fe_SA). It possessed excellent ORR performance in both acidic and alkali environments. On the one hand, the single atomic structure of active FeN₄ increased the number of active sites, while the highly conductive carbon framework aggrandized the specific surface area so that numerous active sites were fully exposed. The mass activity (0.51 A mg⁻¹) was 4.6 times that of the commercial Pt/C catalyst (0.11 A mg⁻¹). On the other hand, the strong coupling effect between FeN₄ and Pt nanoparticles can anchor Pt nanoparticles during the catalytic process to avoid agglomeration and deactivation. The current did not decrease obviously after 5 k cycles, suggesting dramatic stability. Furthermore, anchoring the metal directly on the surface of the Pt single atom to form a metal-supported monoatomic catalyst is a more novel method, which can achieve higher ORR activity by changing the geometry and electronic structure of the catalyst. Liu et al. [173] prepared the Pt₃M@Pt-SAC catalyst with a core–shell structure by fixing a PtCo alloy on Pt-SAC. The synergistic effect between the Pt₃Co alloy and Pt single atom changed the surface electronic structure and accelerated the adsorption and desorption of *OOH. The Pt₃Co@Pt-SAC catalyst exhibited excellent mass activity (1.4 A mg_Pt⁻¹), which was an order of magnitude higher than that of the commercial Pt/C catalyst. In the meantime, the core–shell structure of Pt-SAC is capable of avoiding alloy leaching. Thus, it demonstrated high durability, and the half-wave potential attenuation was only 10 mV after 50 k cycles.

Recent studies have revealed that the surface electronic structure of Pt is highly influenced by the coordination environment around it, owing to the simple structure of the Pt single atom [218,219]. Therefore, the Pt single atom can independently undergo 4e⁻ ORR by changing the coordination environment reaction of metal atoms and reducing the adsorption energy of intermediates [218,219,220,221]. Zhao et al. [26] controlled the type of neighboring dopants and the density of the Pt sites to change the local coordination environment of the Pt single atom, thereby regulating the selection pathway. Compared with Pt-N and Pt-S, Pt-C was more conducive to the desorption of oxygen-containing intermediates (*OOH) and increased the ORR performance. In addition, the increase in platinum loading can make the catalyst layer thicker so that the generated H₂O₂ can be converted into H₂O in a longer diffusion path, thus reducing the selectivity of peroxide and realizing the conversion of 2e⁻ ORR to 4e⁻ ORR. On the basis of the Pt-N-C catalyst, Ni et al. [67] adjusted the environment around N coordination atoms to change the electronic structure of the Pt surface so as to realize the precise regulation of the Pt single-atom reaction pathway. The results illustrate that the pyridinic- and pyrrolic-N coordination in the first shell of the Pt-N-C catalyst promoted the 4e⁻ ORR, while the introduction of graphite-N dopant in the second shell weakened the binding strength of *OOH on the Pt surface, which was not conducive to the 4e⁻ pathway. This provided a basis for the synthesis of Pt catalysts with higher selectivity. Chen et al. [222] conducted a study wherein they supported the Pt monoatomic on S-doped graphite carbon nitride (SGCN). The doping of S made the C and N atoms lacked electrons after the charge redistribution in the GCN support and therefore enhanced the interaction between the metal and the support as well as accelerated the desorption process of the oxygen-containing intermediate. Pt/GCN with a Pt loading of 20.6 wt% (20Pt/SGCN-500) exhibited the best ORR performance with a mass activity of 0.68 A mg_Pt⁻¹, which is much higher than that of 20Pt/GCN and the commercial Pt/C catalyst. Additionally, the half-wave potential of the 20Pt/SGCN-500 catalyst only shifted negatively by 8 mV after the 1 k cycle test, revealing high stability, which was due to the anchoring and strong space limitation of SGCN to Pt single atoms.

Even though single-atom Pt catalysts are currently one of the most promising catalysts and have numerous benefits, there are still certain issues that need to be resolved right away. Isolated metal atoms are easy to aggregate and couple to form clusters, resulting in a sharp decrease in activity [223,224,225]. Moreover, when the single-atom Pt catalyst is in harsh environments such as strong acidic, strong alkali, and high-temperature environments, individual metal active sites are easily destroyed and dissolved. Therefore, at this stage, the performance of Pt-based catalysts can be improved in two ways:

• Increasing the load of Pt while ensuring the uniform dispersion of atoms. In this regard, the development of techniques such as the atomic layer deposition, the photochemical reduction, and the wet chemical method can stabilize the defect sites of loaded metal atoms. Therefore, increasing the number of anchoring sites and the density of active sites through defect engineering is the key to single-atom Pt catalyst synthesis [226,227]. Recent studies have shown that Pt nanostructures rich in GB sites synthesized by GB engineering provide the best coordination environment for the reaction and can increase the residence time of oxygen, which is an effective strategy to improve the catalytic performance of ORR catalysts in the future. However, the grain boundary will accelerate the oxidation of Pt and reduce the stability of the catalyst, so further research is needed [41].
• On the basis of clarifying the deactivation mechanism, new supports capable of anchoring single atoms need to be developed to inhibit the change in the electronic structure of the catalyst during the reaction, thereby improving the stability of the catalyst [228,229].

7. Conclusions and Outlook

This paper focuses on the important part of the cathode catalyst layer—the cathode catalyst. Researchers have proposed adding transition metals to pure platinum catalysts in order to prepare alloy catalysts and using the coordination effect between them to change the electronic structures and properties of surface metals, thereby improving the intrinsic activity of catalysts. For the sake of further enhancing the ORR performance, researchers have synthesized ternary alloy catalysts by N doping or adding auxiliary metals on alloy catalysts. Nevertheless, the stability issues with Pt-based alloy catalysts, which are easy to dissolve under harsh conditions such as strong acid-base environments and high overpotential, have not been well solved. In response to these problems, researchers have developed a series of optimization and improvement methods for Pt-based catalysts. The high-temperature annealing method was used to transform the structure of Pt-based intermetallic catalysts from a disordered to an ordered structure. In ordered intermetallic catalysts, the core–shell structure can effectively protect the core of the intermetallic compound and significantly improve the stability of the catalyst, especially with Pt as the shell. However, the active areas of most core–shell intermetallic catalysts need to be improved. On the basis of controlling the structure, doping stable metals or heteroatoms (O, S, or N, etc.) can effectively improve the activity and stability of the catalysts. With more in-depth research, it was found that the degree of order is positively correlated with ORR activity, and that the size and distribution of particles have a great influence on stability. Therefore, improving the ordering degree of the catalyst under the premise of ensuring a small size is the criterion for improving the intrinsic activity and stability of the catalyst. However, since the catalytic mechanism and nature of ordered intermetallic catalysts have not been fully understood, optimizing the ORR performance of catalysts and understanding the reaction mechanism through advanced characterization methods have become urgent research focuses. From the perspective of element composition, researchers have added constituent elements of catalysts to five or more, and the entropy of mixed configuration was increased to greater than 1.5R to obtain a high-entropy alloy catalyst. The unique high-entropy effect, the lattice distortion effect, the sluggish diffusion effect, and the cocktail effect make it achieve better performance, and it is one of the most promising cathode catalysts at present. From the perspective of optimizing the structure, researchers have improved the ORR performance by controlling the crystal geometry and structural ordering and reducing the particle size to the nanometer level. From the angle of controlling composition, researchers have optimized the catalytic performance by accurately regulating the type and composition of elements. The research on high-entropy alloy catalysts is still in its infancy. There are some problems that still need to be further researched, such as the understanding of the reaction mechanism and pathway, the judgment of the active center and other catalyst design issues, and efficient synthesis methods. Moreover, although low-Pt catalysts have achieved excellent ORR performance, the reduction in Pt loading will lead to the loss of active sites and the obstruction of oxygen transfer. Researchers coupled low-Pt catalysts with PGM-free catalysts, and coupled low-Pt and PGM-free catalysts have abundant pore structures, large specific surface areas, and stable configurations, which are conducive to promoting the four-electron ORR; additionally, the coupling effect observably increases the ORR performance of the catalysts. However, the coupling mechanism and the number of active sites still need to be investigated. Additionally, improving the utilization rate of atoms is another way to achieve the low-cost and high-efficiency catalysis of Pt-based catalysts. Researchers have prepared single-atom Pt catalysts by adjusting the particle size and morphology. The catalysts have excellent performance and have received extensive attention because of their high atomic utilization, rich unsaturated coordination sites, and unique electronic structure. Although Pt single atoms show high selectivity for two-electron ORR, it was found that the addition of Pt single atoms can promote the four-electron ORR of traditional Pt-based catalysts, which provides a new idea for improving the performance of catalysts. In order to realize that single-atom Pt catalysts can complete the four-electron catalytic reaction independently, the researchers optimized the surface electronic structure by changing the coordination environment of the Pt single atom. Nonetheless, the issue of single-atom Pt catalysts readily aggregating has not been solved. It is necessary to further improve the ability of the support to disperse and anchor Pt-catalyzed nanoparticles.

Although a number of Pt-based catalysts with excellent performance have been developed after years of research, there are still some problems to be solved: the structure and composition of the catalyst will deviate from the optimal state during the reaction, especially the alloy catalyst, which will lead to a sharp decline in performance. In addition to the stability strategy proposed from the perspective of inhibiting catalyst dissolution and sintering, more supporting carbon materials can be developed from the perspective of carbon support corrosion to anchor the agglomeration and sintering of metal nanoparticles. With the continuous development of technology, advanced artificial intelligence can be used to design catalysts and carbon materials. For some complex catalysts, such as high-entropy alloy catalysts and coupling catalysts, the mechanisms of action have not been fully elucidated. Therefore, advanced artificial intelligence such as big data and machine learning can be used to explore the relationship between structure and performance and then build a database. In addition, some advanced in situ/operational characterization methods, such as in situ infrared spectroscopy, operando TEM, and SEM techniques, can be used to further explore the reaction mechanism and pathway of the catalyst. This can greatly reduce the trial-and-error cost and help achieve the control and prediction of catalyst performance accurately. However, in the design of catalysts to reduce the loading of Pt while maximizing the utilization of Pt, some new problems may arise; for example, the excellent RDE performance of Pt-based catalysts is difficult to reproduce in MEA. In the future, a number of new test techniques need to be developed to evaluate the activity of the catalysts under high-current-density conditions and simulated real fuel cell conditions so as to narrow the gap between MEA and RDE as much as possible. At the same time, the interaction between MEA components also poses more challenges to PEMFCs. In the future, it will be important to analyze the performance results and design the scheme according to the different components of MEA so as to improve the intersection and interaction of the research fields of MEA components and realize efficient quality transmission.