1 Introduction
With the increasing depletion of global fossil fuel resources and the increasingly serious environmental problems caused by the burning of a large number of fossil fuels, the search and development of new energy and new energy technologies has become one of the most important research topics in the world. Fuel cell is a device that converts the chemical energy of fuel into electrical energy directly, efficiently and environmentally friendly. It has the important advantages of high energy conversion efficiency and environmental friendliness
[1]. Fuel cell technology is known as the most important new energy technology in the 21st century. Fuel cell is mainly composed of electrolyte, anode and cathode. Fuel (hydrogen, alcohols, etc.) Is oxidized at the anode; Air or pure Oxygen undergoes an Oxygen reduction reaction (ORR) at the cathode. Fuel cells can be divided into alkaline fuel cells, phosphoric acid fuel cells, molten carbonate fuel cells, solid oxide fuel cells and proton exchange membrane fuel cells according to the types of electrolytes. Among them, proton exchange membrane fuel cell is considered to be the most likely fuel cell to achieve large-scale commercial application in the short term. At present, proton exchange membrane fuel cell vehicles have developed rapidly in China, Japan, South Korea, the United States and other countries. However, the related problems caused by the slow kinetics of cathode ORR have become the key problems restricting the development of proton exchange membrane fuel cell vehicle industry
[2,3].
For proton exchange membrane fuel cells, the widely used ORR catalyst is Pt/C
[4]. However, Pt is a rare precious metal with limited reserves and high price on the earth, which results in the high cost of fuel cells. Statistics show that the cost of Pt catalyst accounts for about 40% of the total cost of fuel cells, which seriously hinders the commercial development of proton exchange membrane fuel cells. On the other hand, the stability of Pt/C is poor, and the problems of Pt nanoparticles agglomeration and dissolution will occur after long-term use
[5]. It is of great significance to explore low-Pt catalysts with high activity and stability for the large-scale commercialization of proton exchange membrane fuel cells. In this context, researchers at home and abroad have carried out a lot of research work on low Pt catalysts in recent years.
Pt-based alloy catalysts formed by Pt and transition metal M (M = Fe, Co, Ni, Cu, Pd, etc.) Are a kind of very important low-Pt catalysts. Although these catalysts have good activity, there are some problems such as easy dissolution of transition metal and poor stability under harsh conditions such as oxygen-rich cathode, high electrode potential and strong acidic electrolyte in proton exchange membrane fuel cell
[6]. Therefore, in recent years, the study of Pt-based intermetallic compound catalysts with higher activity and stability has become a hot research topic in the field of proton exchange membrane fuel cells.
An intermetallic compound is a compound composed of two or more metallic elements or metallic elements and metalloid elements in an integral atomic ratio. In general, the atoms that make up an intermetallic compound are orderly arranged in two or more sublattices, forming an ordered structure, see Fig. 1. The bonding modes of atoms in intermetallic compounds include metallic, ionic and covalent bonds. Intermetallic compounds have been recognized for a long time because of their structural and bonding characteristics, which make them completely different from their constituent metals. Compared with disordered alloys, intermetallic compounds have ordered structure and strict stoichiometric ratio of elements, and show unique physicochemical properties.
In recent years, researchers at home and abroad have carried out a lot of research work on intermetallic compound catalysts. In this paper, the catalytic mechanism, preparation technology, composition control, particle size control, morphology control and crystal structure of Pt-based intermetallic compound catalysts are discussed.This paper briefly introduces the related research work in recent years, as well as the research progress of Pt-based intermetallic compounds for the cathode oxygen reduction reaction of proton exchange membrane fuel cell, and puts forward some prospects for its challenges and development.
2 Study on the Catalytic Mechanism of Pt-based Intermetallic Compound
As for the catalytic mechanism of intermetallic compounds, Nørskov et al. Proposed a theoretical model of d-band center, and they believed that the ORR activity depends on the adsorption strength between the catalyst surface and the oxygen-containing intermediate
[7]. The adsorption energy of the oxygenated intermediate on the catalyst surface is mainly related to the d-band center of the metal
[8]. Fig. 2 shows the relationship between ORR activity and adsorption energy of oxygen-containing intermediates for different metals
[9]. It can be seen from the figure that the adsorption strength of metal Pt on oxygen-containing intermediates is still too strong, because the d-band center of metal Pt is close to the Fermi level, so the adsorption of oxygen-containing intermediates on the surface of metal Pt is too strong at this level
[10]. The theoretical calculation shows that the catalyst with the binding energy of the oxygen-containing intermediate about 0. 2 eV less than that of metal Pt has the best ORR activity
[9,11]. Therefore, it is an important way to improve the ORR activity of the catalyst by properly reducing the d-band center of Pt and deviating it from the Fermi level negatively.
Some researchers have introduced transition metals into Pt catalysts and developed PtM (M = Co, Fe, Ni, Cu, etc.) alloy catalysts [6,12~14].
The results show that in Pt-based alloy catalysts, the ligand effect (leading to charge redistribution) and stress effect caused by transition metal components make the center position of Pt d-band shift negatively, which weakens the bonding strength between the catalyst surface and oxygen-containing intermediates, and improves the ORR performance of the catalysts. Generally speaking, the ligand effect originates from the charge transfer between elements caused by the difference of electronegativity of different metal elements. The strain effect comes from the lattice mismatch between the Pt element and the transition metal element in the catalyst. Although PtM alloy catalysts often show good ORR activity, transition metals are easily dissolved out under harsh conditions such as oxygen-rich cathode, high electrode potential and strong acidic electrolyte in proton exchange membrane fuel cells, resulting in poor stability of these catalysts [15~17].
The results show that the Pt-based intermetallic compound has better ORR performance and
[18~20] than the Pt-based disordered alloy. The ordered structure of Pt-based intermetallic compounds changes the number and average bond length of Pt-Pt and Pt-M bonds in the catalyst, and strengthens the electronic interaction between Pt and M elements, resulting in a greater degree of lattice mismatch in the intermetallic compounds.The enhancement of coordination effect and strain effect promotes the shift of Pt d-band center, which strongly weakens the adsorption strength between the catalyst surface and oxygen-containing intermediates, so Pt-based intermetallic compounds can further improve the ORR activity. The improved stability of Pt-based intermetallic compounds is mainly due to the following two points. First, the Pt-based intermetallic compound has a more negative enthalpy of formation than the Pt-based disordered alloy and is therefore more stable. Secondly, thermal annealing is the main method to synthesize Pt-based intermetallic compounds at present. Pt is easy to segregate to the surface during thermal annealing to form a core-shell structure with a thickness of several Pt atomic layers. The core-shell structure can effectively prevent the dissolution of internal transition metals and improve stability
[21,22].
Recently, Kim et al. Combined density functional theory (DFT) calculations and experimental results to reveal the reason for the enhanced ORR activity of Pt
3Co intermetallic nanowires (O-Pt
3Co NWs)
[23]. Because the crystal structure of the Pt
3Co intermetallic compound (O-Pt
3Co) and the Pt
3Co disordered alloy (D-Pt
3Co) are both face-centered cubic, they believe that the strength of the strain effect is basically the same. The difference between the binding energies of the O-Pt
3Co and D-Pt
3Co surfaces and the pure Pt (111) surface for the oxygenated intermediate, ΔE
O, was first calculated by DFT, respectively. The results show that the ΔE
O of O-Pt
3Co and D-Pt
3Co are 0.259 eV and 0.278 eV, respectively, see Fig. 3 (d). Based on the current consensus that the catalyst with a ΔE
O of 0.2 eV has the best ORR activity, it is predicted that the ORR activity of O-Pt
3Co is stronger than that of D-Pt
3Co
[24]. In addition, they pointed out that compared with the D-Pt
3Co, the Co arrangement in the subsurface layer of the O-Pt
3Co changes the bonding strength of the d orbital between Pt and Co, so the ΔE
O of the O-Pt
3Co is closer to the optimal value. For experimental verification, free-standing Pt
3Co intermetallic nanowires (O-Pt
3Co NWs) and Pt
3Co alloy nanowires (D-Pt
3Co NWs) with almost the same composition and shape were prepared and subjected to ORR performance test. At 0.9 V (vs. RHE), the ORR mass activity of vs. RHE was 0.31 A/mg
Pt, whereas that of D-Pt
3Co NWs was only 0.15 A/mg
Pt, confirming the theoretical prediction.
The ORR mass activity of the PtCo
3 intermetallic compound (O-PtCo
3/C) prepared by Wang et al. Was 6 and 1.3 times higher than that of the commercial Pt/C and PtCo
3 disordered alloy (D-PtCo
3/C), respectively
[25]. They calculated the d-band center position of pure Pt, O-PtCo
3/C and D-PtCo
3/C by DFT to reveal the reason for the enhancement of O-PtCo
3/C ORR activity. The calculated results show that the d-band center of pure Pt is located at -2.2 eV. The d-band center of D-PtCo
3/C after Pt alloying with Co is located at − 2.267 eV, while the d-band center of O-PtCo
3/C after D-PtCo
3/C ordering is located at − 2.32 eV. The results show that the d-band center of the O-PtCo
3/C is closer to the optimal position than that of the D-PtCo
3/C, so the ORR activity is enhanced. They believed that the near-optimal value of the d-band center of O-PtCo
3/C was attributed to its stronger strain effect, which was beneficial to optimize the adsorption strength of Pt on oxygen-containing intermediates and the dissociation of O — O bonds.
3 Preparation Technology and Its Research Progress
It is difficult to prepare intermetallic compounds, especially nanoscale intermetallic compounds, because of the highly ordered arrangement of atoms and the high diffusion barrier of atoms. Based on the rapid development of colloidal interface chemistry, the main methods to synthesize nanoscale intermetallic compounds are thermal annealing, wet chemistry and chemical vapor deposition.
3.1 Thermal annealing
Thermal annealing is a common method to synthesize nanoscale intermetallic compounds, which can be divided into two main methods. One is to synthesize disordered alloys by solvothermal method and organic liquid phase reduction method, and then anneal the disordered alloys in inert or reducing atmosphere to induce the formation of ordered structures. The other is a method in which the metal precursor is directly impregnated into the support and then thermally annealed in a reducing atmosphere. Wang et al. Reported the preparation of Pt
3Co disordered alloy nanoparticles using metal acetylacetonate as precursor, tetradecanediol as reducing agent, oleylamine and oleic acid as stabilizers, and dichlorobenzene and benzyl ether as mixed solvent at 200 ° C for 1 H
[26]. The Pt
3Co intermetallic compound was obtained after further treatment of the Pt
3Co disordered alloy nanoparticles at 600 ° C for 4 H. Chen et al. Dissolved H
2PtCl
6·6H
2O and NiCl
2·6H
2O in a proper amount of water, impregnated them on XC-72 carbon, and reduced them at 300 ℃ for 2 H in hydrogen atmosphere
[27]. In order to induce the formation of ordered structure, Pt
3Ni
2/C intermetallic compound supported on XC-72 carbon was obtained by thermal annealing at 600 ℃.
3.2 Wet chemical method
Unfortunately, thermal annealing can hardly control the nanostructure of intermetallic compounds, and the obtained nanocrystals are all irregular polyhedrons, which usually greatly affect the ORR performance of materials. In fact, the low-temperature wet chemical method may be one of the best methods to synthesize intermetallic nanocrystals with specific nanostructures
[28]. The wet chemical method is mainly based on the continuous or even simultaneous occurrence of the two processes of disordered alloy formation and disorder-order phase transformation, and the synthesis temperature is relatively low (< 300 ℃)
[29]. Because the rapid reduction of different metal precursors usually leads to the formation of solid solution alloys, weak reducing agents are often used at low temperatures in wet chemical synthesis to achieve the co-reduction of different metal precursors and the disorder-order phase transition successively or even simultaneously, thus promoting the formation of ordered structures. In addition, the intermetallic compound with special morphology can be prepared by the method by selecting a proper capping agent, a stabilizer and the like. The capping agent is selectively adsorbed on some special crystal planes, which plays a role in regulating the growth direction of nanocrystals, reducing the surface free energy of nanocrystals, and promoting the disorder-order phase transition. Bu et al. Reduced Pt(acac)
2 and Pb(acac)
2 at 160 ℃ with octadecene and oleylamine as mixed solvent and ascorbic acid as reducing agent, and obtained PtPb intermetallic compound with hexagonal plate shape
[30]. Luo et al. Obtained PtSnBi ordered intermetallic nanoplates by reducing Pt(acac)
2, SnCl
2, and Bi(act)
3 at 220 ° C with octadecene and oleylamine as mixed solvents and ascorbic acid as a reducing agent
[31]. Rong et al. Reduced Pt(acac)
2 and SnCl
2 with polyvinylpyrrolidone as a stabilizer, N, N-dimethylformamide as a solvent and a reducing agent to obtain cube-shaped PtSn
3 intermetallic nanocrystals
[32]. In addition, they increased the precursor concentration by two times to obtain concave cubic PtSn
3 intermetallic nanocrystals, and by eight times to obtain defect-rich PtSn
3 intermetallic cubic nanocrystals. In this reaction system, the reduction rate of Pt is higher than that of Sn. N, N-dimethylformamide is a suitable solvent for the reduction of Pt and Sn with different rates. If the solvent is replaced by benzyl alcohol, which has stronger reducibility, only PtSn
3 intermetallic nanocrystals with common shape can be obtained.
If the standard redox potentials of the metal precursors in the system are quite different, it will be a great challenge to synthesize Pt-based intermetallic compounds by using mild reducing agents, while the use of strong reducing agents is not conducive to the morphology control of nanocrystals. In addition, the wet chemical method will produce a large amount of waste liquid, which is likely to cause environmental pollution in industrial production.
3.3 Chemical vapor deposition
Chemical vapor deposition (CVD) is a process that uses gaseous or vapor-like substances to react chemically on a solid surface to form a solid deposit. In short, CVD is a process in which a variety of gaseous raw materials are introduced into a reaction chamber to react with each other to form new materials and then deposited on the surface of the substrate. In recent years, some researchers have used CVD to prepare Pt-based intermetallic compounds. In the CVD process, Pt nanocrystals supported on an inert support are often used as the matrix, and the second component is selectively deposited on the Pt surface. Saedy et al. Used a fluidized CVD reactor to selectively deposit Co on the surface of ultra-small Pt nanocrystals supported on cerium oxide nanorods to prepare Pt
3Co intermetallic compounds with an average particle size of 1.1 nm
[33]. However, it is difficult to precisely control the final structure and composition of the material by selective deposition of non-noble metals, so it is difficult to prepare Pt-based intermetallic compounds for ORR by this method.
Generally speaking, CVD method is difficult to accurately control the structure and composition of Pt-based intermetallic compounds, and it is difficult to become the mainstream synthesis method in the future. Secondly, although the wet chemical method can be used to synthesize Pt-based intermetallic compounds with special morphologies, it has few applicable systems, poor repeatability and produces more waste liquid. Therefore, in the future, we can focus on the synthesis mechanism of wet chemical method and seek alternative and recyclable clean solvents, laying a solid foundation for large-scale and industrial production of Pt-based intermetallic compounds. Finally, for a long time in the future, annealing is an important method for the synthesis of Pt-based intermetallic compounds, but it needs to continue to focus on solving the problems of nanocrystal agglomeration, uneven size distribution and irregular morphology caused by high temperature.
4 Research Progress of Composition Regulation
At present, in order to improve the ORR activity, Pt and transition metals or metalloids are often formed into alloys or intermetallic compounds. Other components can reorganize the d-electron orbital of Pt, shift the d-band center of Pt, reduce the binding energy between Pt and oxygen-containing intermediates, and improve the ORR activity
[34].
4.1 Binary Pt-based intermetallic compound
The Pt-based intermetallic compound has a strict stoichiometric ratio between the constituent elements, and different atoms occupy specific sites of the lattice, so it can only be formed in a certain range of composition and temperature. At present, the phase diagram can be used to predict the formation of intermetallic compounds. The atomic radius and electronegativity of elements also have an effect on the formation of intermetallic compounds. According to the Hume-Rothery law, a binary intermetallic compound can be formed only if the difference in atomic radius between the two elements is less than 15%
[35]. In addition, the difference of electronegativity between elements is beneficial to the formation of ionic bonds and covalent bonds, which promotes the formation of ordered structures. A large number of studies have shown that Pt can form binary Pt-based intermetallic compounds with transition metals such as Fe, Co, Ni and Cu. Transition metals rearrange the d-electron orbital of Pt, shift the d-band center, and improve the ORR activity. Zou et al. Prepared PtFe intermetallic compound catalyst (PtFe/C @ NC-3-800)
[36]. The mass activity of PtFe/C @ NC-3-800 at 0.9 V (vs. RHE) is vs. RHE, which is 6.73 times that of Pt/C(0.22 A/mg
Pt). X-ray photoelectron spectroscopy results show that the Pt 4F binding energy of PtFe/C @ NC-3-800 is negatively shifted to 0. 71 eV compared with that of Pt/C, which proves that Pt in PtFe/C @ NC-3-800 receives electrons from Fe, and the d-band center is negatively shifted, which weakens the binding energy between Pt and oxygen-containing intermediates and improves the ORR activity. The mass activity of PtFe/C @ NC-3-800 decreased by only 20.64% after 30 000 cycles of cyclic voltammetry, and they attributed the excellent stability of PtFe/C @ NC-3-800 to the ordered structure.
4.2 Ternary Pt-based intermetallic compound
In recent years, researchers have found that the ORR performance of the catalyst can be greatly improved by introducing a third component into the binary Pt-based intermetallic compound, because the third component can finely control the ordered structure, optimize the strain, coordination and overall effect. Zhu et al. Reported that the introduction of a small amount of Cu into the Pt-Fe system can induce the formation of ordered structure, and PtFe
0.9Cu
0.1 intermetallic compounds can be obtained by thermal annealing at 500 ℃, and their ORR properties are significantly improved, which is the result of the combined effect of ordered structure and strain effect caused by Cu
[37]. Liang et al. Doped W, Ga and Zn into L1
0-PtCo/C intermetallic compounds to prepare L1
0-M-PtCo/C(M=W, Ga, Zn) intermetallic compounds, respectively
[38]. Among them, L1
0-W-PtCo/C showed the best ORR activity. The extended X-ray absorption fine structure results show a d
Pt-Pt of 2.765 765 Å for the Pt foil and a d
Pt-Pt of 2.686 686 Å for the A1-PtCo disordered alloy, while the d
Pt-Pt slightly grows to 2.707 707 Å after performing ordering and W doping. The slightly reduced d
Pt-Pt of L1
0-W-PtCo/C weakens the oxygen adsorption energy and increases the ORR activity. This work provides a reference for the synthesis of ternary Pt-based intermetallic compounds.
The results show that the ORR performance can also be improved by introducing non-metallic elements into binary Pt-based intermetallic compounds. Liu et al. Prepared N-doped vs. RHE intermetallic compound catalyst (O-Pt
3CoN/C), and its mass activity at 0.9 V (vs. RHE) was 1.2 and 2.4 times higher than that of N-undoped vs. RHE intermetallic compound catalyst and commercial Pt/C, respectively
[39]. After 20 000 cycles of cyclic voltammetry, only 23.1% of the mass activity was lost. Zhao et al. Synthesized N-doped PtNiN intermetallic compound catalyst (Int-PtNiN/KB)
[40]. Int-PtNiN/KB exhibited excellent ORR performance with an ORR mass activity of 1.83 A/mg
Pt, which is 10.1 times higher than that of commercial Pt/C. After 30 000 cycles of cyclic voltammetry, the mass activity of Int-PtNiN/KB lost only 27.3%, while the commercial Pt/C lost as much as 51.1%. They confirmed by DFT calculations that the ordered structure and the tensile strain caused by N doping optimized the binding strength of Pt with oxygen-containing intermediates and improved the electron transfer rate, and the ORR activity was enhanced. In addition, Int-PtNiN/KB has a unique ordered structure with alternating Ni
4-N planes and Pt planes. They attributed the excellent stability of Int-PtNiN/KB to the unique ordered structure of Int-PtNiN/KB effectively preventing the dissolution of Ni. In this study, thermal annealing was carried out in NH
3 atmosphere, and the disorder-order phase transition and N doping were realized at the same time, which is a relatively simple synthesis method and provides a reference way for further improving the ORR activity and durability of PtM catalysts. However, due to the differences in the physical and chemical properties of the elements, the strain effect and coordination effect of the third component on Pt are also different. In the future, the fine control of Pt-based intermetallic compounds can be achieved by changing the type and addition of the third component.
5 Research Progress of Granularity Regulation
Nanocrystalline particle size has an important effect on the activity of supported metal nanocatalysts. From the perspective of geometric structure, with the decrease of nanocrystal size, more low-coordinated atoms are exposed, which can significantly change the structure and proportion of catalyst active centers
[41,42]. From the perspective of electronic structure, the electronic energy levels of metal particles will change significantly due to the quantum size effect, which greatly affects the orbital hybridization and charge transfer between the catalyst active center and the reactant
[43,44].
At present, thermal annealing is still the main method to synthesize intermetallic compounds
[45,46]. To achieve the disorder-order phase transition, the thermal annealing temperature is usually higher than 500 ° C
[47]. However, high temperature can easily lead to agglomeration and sintering of intermetallic compound nanocrystals, resulting in larger particle size. At present, it is still difficult to synthesize intermetallic compounds below 5 nm. Reducing the particle size of the intermetallic compound can expose more active sites of the nanocrystal, improve the utilization rate of Pt, adjust the geometric structure and electronic structure of the intermetallic compound, and improve the ORR activity. At present, the main methods to control the particle size of intermetallic compounds are coating method, KCl matrix method and space confinement method.
5.1 Coating method
Coating method refers to the method of coating a sacrificial protective layer on the surface of disordered alloy nanoparticles to prevent the migration, agglomeration and sintering of nanocrystals during thermal annealing. Kim et al. Obtained L1
0-fct PtFe intermetallic compounds with smaller particle size by coating nanocrystals of fcc-PtFe disordered alloy with MgO and thermal annealing at 700 ℃
[48]. During the heat treatment at high temperature, the MgO protective layer on the surface of the nanocrystals effectively prevents the agglomeration and sintering of PtFe nanocrystals, but the MgO protective layer restricts the diffusion and migration of atoms in the nanocrystals, and only partially ordered L1
0-fct PtFe intermetallic compounds can be obtained. After thermal annealing, the oxide protective layer on the surface of the nanocrystal needs to be removed by pickling and other methods, and the process is cumbersome. Therefore, researchers developed the carbon coating method. Chung et al. Obtained a PtFe intermetallic compound catalyst (fct-PtFe/C) with an average particle size of 6.5 nm by coating polydopamine on the surface of fcc-PtFe disordered alloy nanocrystals and thermal annealing at 700 ℃
[49]. The fcc-PtFe nanocrystals without polydopamine coating exhibit severe agglomeration and sintering, and the particle size can reach tens of nanometers, which indicates that carbon coating can effectively prevent the agglomeration and sintering of nanoparticles. Jung et al. Used 1,2-hexadecanediol and iron acetylacetonate coated Pt-Fe disordered alloy catalyst (Pt-Fe/C) followed by thermal annealing to obtain Pt
3Fe intermetallic compound catalyst (Pt
3Fe/C)
[34]. Thermal annealing induces a disorder-order phase transition and removes the unnecessary carbon layer. The carbon layer derived from the coated 1,2-hexadecanediol and iron acetylacetonate during annealing acts as a good barrier, and the average particle size of the coated Pt-Fe/C is 6.77 nm after thermal annealing. However, the uncoated Pt-Fe/C has serious agglomeration after thermal annealing, and the average particle size is about 10. 74 nm. The ORR mass activity of Pt
3Fe/C is 0.454 A/mg
Pt, which is four times higher than that of commercial Pt/C. After 8000 cycles of accelerated durability test, the mass activity of Pt
3Fe/C was 0.343 A/mg
Pt, while that of Pt-Fe/C was only 0.109 A/mg<sub>Pt. </sub>
5.2 KCl matrix method
The KCl matrix method refers to the KCl as a by-product of the synthesis system, which acts as a physical barrier to hinder the migration and agglomeration of nanocrystals during thermal annealing, as shown in Figure 4. After thermal annealing, KCl can be removed by water washing and other means to obtain intermetallic compound nanocrystals. Chen et al. Proposed that the particle size of the intermetallic compound can be effectively controlled by changing the molar ratio of the metal precursor to KCl and the annealing temperature and time
[50].
Cui et al. Prepared Pt
3Cr disordered alloy nanoparticles trapped in KCl matrix, which were thermally annealed and then washed away the KCl matrix with ethylene glycol to obtain a Pt
3Cr intermetallic compound catalyst (Pt
3Cr/C) with a particle size of about 5 nm
[51]. The ORR mass activity of vs. RHE at 0.9 V (vs. RHE) is 237 mA/mg
Pt, which is about 2.3 times higher than that of homemade Pt/C. More importantly, the mass activity loss of Pt
3Cr/C was only 14.7% after 5000 cycles of cyclic voltammetry test, and the leaching loss of Cr was only 13.5% after 4 weeks of test. This work provides guidance for optimizing the synthesis method of intermetallic compound catalysts and improving their ORR performance. Controlling the particle size of intermetallic compounds by KCl matrix can reduce the use of surfactants and is environmentally friendly.
5.3 Space limited region method
Space confinement method is a method to confine the agglomeration and sintering of nanocrystals in space during annealing by using materials with space structure as carriers. Zhao et al. Prepared Pt
3Co/C and Pt
3Co/DMC intermetallic compound catalysts using Vulcan XC-72 carbon and ZIF-8 derived carbon with mesoporous structure as supports, respectively
[52]. The average particle size of Pt
3Co/C is 6.6 nm, while the average particle size of Pt
3Co/DMC is only 3 nm, and the particle size distribution is narrow, which indicates that the three-dimensional pore structure of ZIF-8-derived carbon effectively restricts the agglomeration and sintering of Pt
3Co nanocrystals during annealing. The oxygen reduction mass activity of Pt
3Co/DMC can reach 0.83 A/mg
Pt, which is 4.6 and 5.2 times higher than that of Pt
3Co/C and commercial Pt/C, respectively. They attributed the excellent mass activity of Pt
3Co/DMC to its smaller particle size, which exposes more active sites. The mass activity of the Pt
3Co/DMC decreased by only 13.3% after 5000 cycles of accelerated durability test, which benefited from the ordered structure and strong metal-support interaction. Although the spatial confinement method can restrict metal sintering by means of spatial structure, it will also increase the diffusion resistance of reactants and products, which may lead to difficulties in water management in the practical application of proton exchange membrane fuel cells.
6 Research Progress of Morphology Control
The morphology of the catalyst is directly related to the exposed crystal plane, which affects the mass transfer and the interaction between the metal and the support
[53]. Because ORR is a surface reaction, the morphology of the catalyst has an important influence on the ORR performance. A large number of studies have shown that the ORR activity of Pt-based disordered alloy catalysts with special morphology is very superior
[54,55]. However, the chemical stability and electrochemical durability are very poor, which can not meet the practical application needs of proton exchange membrane fuel cells
[56,57]. Therefore, people have turned their attention to Pt-based intermetallic compound catalysts with better durability. In recent years, some researchers have synthesized Pt-based intermetallic compound catalysts with special morphologies, and the reported special morphologies include: core-shell structure, nanoplate
[30,59~61], nanowire and nanoframe, and these catalysts usually show better ORR performance
[21,58][60][47,62].
6.1 Core-shell structure
The ordered assembly structure formed by one nanomaterial coating another nanomaterial through chemical bonds or other forces is the core-shell structure. At present, the core-shell structure can be constructed by means of surface thermal segregation, electrochemical deposition or dealloying, and a Pt shell with a thickness of several atomic layers can be formed on the surface of intermetallic compounds. Due to the lattice mismatch between the core and the shell, the intermetallic core produces a strain effect on the surface Pt shell, which improves the ORR activity
[63]. Secondly, the Pt shell on the surface can prevent the dissolution of the transition metal in the core, and the durability of the catalyst is greatly improved
[64].
Cheng et al. Prepared Cu
3Pt alloy catalyst (Pt-D-Cu
3Pt/C) and intermetallic compound catalyst (Pt-O-Cu
3Pt/C) with core-shell structure by electro-dealloying method
[58]. The ORR specific activity of Pt-O-Cu
3Pt/C was 1.4-fold higher than that of Pt-D-Cu
3Pt/C. The DFT theoretical calculation shows that the intermetallic core of Pt-O-Cu
3Pt/C produces more significant compressive strain on the surface Pt shell, which can further reduce the d-band center position of the Pt shell and improve the ORR activity. Xiong et al. Synthesized a Pt
3Co intermetallic compound catalyst (Pt
3Co/C) with a core-shell structure by thermal annealing, and its Pt shell thickness is about 0.6 nm, which is about 2 – 3 Pt atomic layers
[65]. The half-wave potential of vs. RHE with a total metal loading of 40% was as high as 0.943 V (vs. RHE), which was 17 mV higher than that of vs. RHE with a loading of 20%. After 4000 cycles of cyclic voltammetry, 97% of the electrochemical specific surface area was retained. Such superior ORR performance is mainly attributed to the Pt shell on the surface of Pt-O-Cu
3Pt/C and the stable intermetallic core of Pt
3Co.
6.2 One-dimensional nanowire
Compared with zero-dimensional nanoparticles, one-dimensional nanowires (NWs) have special physical and chemical properties. Nanowires have anisotropic structure, high specific surface area and strong conductivity, and are rich in high-index crystal planes and defects, which are beneficial to improving ORR activity
[66,67]. Secondly, compared with zero-dimensional nanoparticles, the contact sites between nanowires and the carrier are relatively more, and the nanowire crystals can be firmly anchored on the carrier, effectively preventing the agglomeration and sintering of nanocrystals. In collaboration with Guo Shaojun of Peking University, Huang Xiaoqing et al. Of Soochow University synthesized a series of Pt-Co intermetallic nanowires with high-index and Pt-rich facets by wet chemical method
[67]. The ORR activity of the Pt
3Co intermetallic nanoparticle catalyst is much lower than that of the Pt
3Co intermetallic nanowire catalyst. The mass activity of the vs. RHE intermetallic nanowire catalyst at 0.9 V (vs. RHE) is 33.7 times that of commercial Pt/C. At the same time, the Pt
3Co ordered intermetallic nanowire catalyst showed excellent stability, retaining 91. 9% of the mass activity after 20 000 cycles of cyclic voltammetry. Kim et al. Synthesized free-standing Pt
3Co alloy nanowires (D-Pt
3Co NWs) using mesoporous silica as a template, and obtained free-standing Pt
3Co intermetallic nanowires (O-Pt
3Co NWs) after thermal annealing
[23]. At 0.9 V (vs. RHE), the ORR mass activity of vs. RHE was 0.31 A/mg
Pt, whereas that of D-Pt
3Co NWs was only 0.15 A/mg
Pt. This O-Pt
3Co NWs is the best one among the reported Pt-based free-standing ORR catalysts.
6.3 Two-dimensional nanoplate
Two-dimensional nanoplates (NPs) have a relatively high proportion of surface atoms and more exposed active sites, which is conducive to improving ORR activity
[68]. Secondly, the interaction between the nanoplate and the support is strong, which is beneficial to improve the long-range stability of the catalyst. The organic combination of ordered structure and two-dimensional nanoplates can greatly improve the ORR performance, so it has attracted wide attention. Guo et al. Synthesized PtBi intermetallic nanoplates by wet chemical method, and high-angle annular dark-field scanning transmission electron microscopy and transmission electron microscopy confirmed that the prepared PtBi nanocrystals were hexagonal nanoplates with a size of about 100 nm and a thickness of about 18 nm
[69]. They pointed out that the PtBi intermetallic nanoplate has excellent resistance to CH
3OH, and the current density of the PtBi intermetallic nanoplate can be maintained at a large value even in a 0.5 mol/L CH
3OH solution. The mass activity of PtBi intermetallic nanoplates is vs. RHE at 0.85 V (vs. RHE), which is 4.7 times higher than that of PtBi intermetallic nanoparticles. Chronoamperometric stability test results show that PtBi intermetallic nanoplates have excellent tolerance to CH
3OH, HCOOH, and CO relative to commercial Pt/C. They suggested that the charge transfer caused by the Pt-Bi bond may help to improve the tolerance of the catalyst to small molecular organic fuels. This work opens up a new way to create new high-performance ORR electrocatalysts and provides a reference way to improve the stability and tolerance of cathode catalysts for proton exchange membrane fuel cells.
They also reported the one-pot synthesis of a series of hcp-PtBi/fcc-Pt intermetallic nanoplates with core-shell structure, whose ORR activity was 5 times higher than that of commercial Pt/C
[59]. The results of X-ray photoelectron spectroscopy and DFT calculation show that the edge of Pt shell and PtBi intermetallic core of hcp-PtBi/fcc-Pt intermetallic nanoplates has the optimal oxygen adsorption energy due to the Bi-p empty band suppression at the core/shell interface, which is the reason for the enhancement of ORR activity. Thanks to the ordered structure and core/shell structure, the hcp-PtBi/fcc-Pt intermetallic nanoplates showed almost no loss of ORR activity in the accelerated durability test. This work provides a reference for the rational design of novel two-dimensional core/shell nanostructured high-performance catalysts for ORR.
6.4 3D nanoframe
The three-dimensional structure of three-dimensional Nanoframes (NFs) is beneficial to mass transfer, and the high specific surface area can maximize the efficiency of Pt utilization. However, the very thin framework in three-dimensional disordered nanoframes leads to poor chemical stability and ORR durability, so Pt-based intermetallic nanoframes have attracted attention
[70,71]. Kim et al. Reported the L1
1-PtCu intermetallic nanoframe catalyst (o-PtCuNF/C)
[47]. The aggregation of PtCu disordered nanoframes (D-PtCuNF/C) during annealing was inhibited by silica coating, and the rhombic dodecahedron morphology was maintained. The ORR specific activity of o-PtCuNF/C was 4.69 mA/C
, which was 2.2 times that of d-PtCuNF/C and 14.2 times that of commercial Pt/C. Importantly, the ORR mass activity of o-PtCuNF/C exceeds that of PtCu-based catalysts reported so far. After 10 000 cycles of accelerated durability test, the specific activity of O-PtCuNF/C remained 89.1%, while the specific activity of D-PtCuNF/C decreased by 38.2%. They attributed the excellent stability of o-PtCuNF/C to the higher chemical stability of the ordered structure itself. Gong et al. Synthesized L1
1-PtCu intermetallic nanoframes (L1
1-PtCu NFs) by thermal annealing of PtCu disordered nanoframes (PtCu NFs)
[62]. The mass activity of PtCu NFs is 0.82 A/mg
Pt, while the mass activity of L1
1-PtCu NFs is only 0.44 A/mg
Pt, because the nanocrystals agglomerate during thermal annealing. However, the durability of L1
1-PtCu NFs is greatly improved, and its mass activity is only 15% after 30 000 cycles of cyclic voltammetry, while the mass activity of PtCu NFs is 70% after 10 000 cycles of cyclic voltammetry. They attribute the excellent stability of L1
1-PtCu NFs to the ordered structure which effectively prevents the dissolution of Cu. In-situ heating experiments under scanning transmission electron microscope (SEM) show that thermal annealing at appropriate temperature can reduce the barrier of atom migration and accelerate atom diffusion. Thermal annealing at higher temperatures can lead to excessive atom migration, resulting in the collapse and aggregation of nanocrystalline structures, so the synthesis of morphology-controlled intermetallic compounds is still a great challenge.
At present, the main way to synthesize Pt-based intermetallic compounds is still thermal annealing, and the thermal annealing temperature is generally higher than 500 ℃
[47]. The rapid surface diffusion at high temperature leads to the change of nanocrystal morphology, and the product is difficult to maintain the initial morphology. At present, the main way to synthesize Pt-based intermetallic compound catalysts with special morphology is wet chemical method. In general, the size of the product obtained by wet chemical method is larger, which reduces the utilization of Pt to a certain extent. Therefore, the synthesis of small Pt-based intermetallic compound catalysts with special morphology is still a challenging research work.
7 Progress in Crystal Structure Research
Compared with disordered alloys, in binary Pt-based intermetallic compounds, Pt and M atoms occupy specific sites of the lattice respectively, and the overall structure is long-range ordered. According to the atomic ratio, binary Pt-based intermetallic compounds can be generally divided into three types: PtM, PtM3, and Pt3M. The crystal structure of binary Pt-based intermetallic compounds is mainly divided into three types: L10, L11 and L12, while the metal Pt is A1 type face-centered cubic crystal. In the L10-type binary Pt-based intermetallic compound, the face-centered cubic lattice is composed of M atomic planes and Pt atomic planes stacked alternately. PtCu intermetallic compound is the only L11-type binary Pt-based intermetallic compound found so far (Fig. 5). After the PtCu alloy is ordered, the (111) plane of the original face-centered cubic is alternately occupied by Cu and Pt atoms, and the crystal structure changes from cubic to rhombohedral, as shown in fig. 5. In the L12-type PtM3 intermetallic compound, Pt atoms occupy eight vertex sites and M occupies six face-centered sites of the lattice, while the atomic arrangement of the L12-type Pt3M intermetallic compound is exactly opposite to the former, see Fig. 1.
In the ternary Pt-based intermetallic compound Pt-M
1-M
2, the lattice sites originally occupied by M atoms are randomly occupied by two kinds of atoms, M
1 and M
2[72]. However, in some special cases, the third component M
2 can occupy the lattice sites originally occupied by Pt atoms. Kim et al. Reported a L1
1-PtCu intermetallic nanoframe (o-PtCuNF/C) with a specific ORR activity of 4.69 mA/
, which is 14.2 times higher than that of the commercial Pt/C catalyst
[47]. Li et al. Prepared L1
0-type FePt intermetallic compound catalyst (L1
0-FePt/Pt)
[73]. The mass activity of L1
0-FePt/Pt is 0.7 A/
L1
0-FePt/Pt has no significant loss of mass activity after 30 000 cycles of cyclic voltammetry. Wang et al. Synthesized L1
2 type PtCo
3 intermetallic compound with an ORR specific activity of 1.74 mA/
at 0.9 V (vs. RHE), while the specific activity of the corresponding A1 type PtCo
3 alloy was only 1.18 mA/
[26]. Zhao et al. Synthesized N-doped L1
1-type PtCu intermetallic compound catalyst (Int-PtCuN/KB) by thermal annealing
[74]. The half-wave potential of Int-PtCuN/KB can reach 916 mV, which is 49 mV higher than that of commercial Pt/C. The doped N provides additional compressive strain and improves the ORR activity. The mass activity of Int-PtCuN/KB at 0.9 V (vs. RHE) was vs. RHE, which was 5.2 times higher than that of commercial Pt/C. The mass activity of Int-PtCuN/KB decreased by only 23.5% after 20 000 cycles of accelerated durability test, which was much lower than that of commercial Pt/C (40.9%). The L1
1 ordered structure and Cu-N bond in Int-PtCuN/KB are the key factors for the stability improvement. Wang et al. Prepared a L1
2-type ordered Pt
3Co intermetallic catalyst (Pt/40Co-NC-900)
[75]. Co atoms occupy six face-centered sites and Pt atoms occupy eight vertex sites in the lattice observed by aberration-corrected high-angle annular dark-field scanning transmission electron microscopy, which, together with the fast Fourier transform results, confirms that Pt/40Co-NC-900 has a L1
2-type ordered structure. The ORR specific activity of Pt/40Co-NC-900 can reach 1.15 mA/vs. RHE at 0.9 V (vs. RHE), and the half-wave potential loss is only 12 mV after 30 000 cycles of cyclic voltammetry, and the excellent stability is attributed to its L1
2-type ordered structure.
8 Conclusion and prospect
As a kind of efficient and stable electrocatalyst, Pt-based intermetallic compound catalyst has attracted more and more attention because of its unique ordered structure, geometric effect, electronic effect and overall effect, which can greatly improve the mass activity of Pt. In this paper, the application of Pt-based intermetallic compounds in the cathode ORR of proton exchange membrane fuel cells was discussed, and the research progress in the preparation technology, composition control, particle size control, morphology control and crystal structure control of Pt-based intermetallic compound catalysts was emphatically discussed. At present, great progress has been made in the research of Pt-based intermetallic compound nanocrystals, but there are still many problems, and it is likely to make breakthroughs in the following aspects in the future.
(1) At present, the research on Pt-based intermetallic compound nanocrystals is mainly focused on Fe, Co, Cu and other elements, and the research on Ni is relatively small. The ORR activity of PtNi disordered alloy catalyst is strong, but its stability is insufficient. One of the feasible ways to improve its stability is to convert PtNi disordered alloy into PtNi intermetallic compound. However, most of the PtNi intermetallic compounds reported so far are thermodynamically stable spheres with large particle size and low ORR activity. Therefore, it is an effective way to improve the ORR catalytic performance of PtNi-based intermetallic compounds by further exploring the preparation methods of PtNi-based intermetallic compounds and precisely controlling the composition/component, particle size, morphology and structure of PtNi-based intermetallic compounds.
(2) Carbon is a typical support material for PEM fuel cell catalysts because of its high specific surface area and high electronic conductivity. However, as a catalyst support for proton exchange membrane fuel cell, carbon faces an important problem-corrosion resistance. The corrosion of carbon support can cause the agglomeration and growth of noble metal catalyst particles and the reduction of active sites. In addition, carbon support corrosion also adversely affects the transport of protons, electrons, and gases, causing irreversible performance losses. Therefore, there is a need to develop supported and/or self-supporting catalysts that are stable at high potentials and high operating temperatures.
(3) A large number of studies have shown that the ORR performance of Pt-based intermetallic compound catalysts is superior to that of Pt-based disordered alloys with the same composition. However, the formation mechanism of intermetallic compounds is still unclear. The study of the formation mechanism of intermetallic compounds is helpful for the simultaneous control of Pt-based intermetallic nanocrystals, such as the simultaneous control of particle size and morphology. Human beings need advanced in-situ characterization techniques to study the formation mechanism of intermetallic compounds, so it is necessary to develop advanced in-situ characterization techniques.
(4) At present, there are many basic studies on Pt-based intermetallic compound catalysts for ORR, but how to combine them with practical applications is also a difficult problem to be solved urgently.
(5) Thermal annealing is an important method for the synthesis of Pt-based intermetallic compound catalysts, but it causes problems such as particle agglomeration and sintering, which limits the improvement of ORR performance and consumes a lot of energy. The synthesis of Pt-based intermetallic compound by wet chemical method can reduce the energy consumption, but it needs more surfactant and has the hidden danger of environmental pollution. Therefore, there is a need to develop new preparation technologies that are low in energy consumption and environmentally friendly.
National Key Research and Development Program of China(2017YFB0102900)
National Key Research and Development Program of China(2016YFB0101201)
National Natural Science Foundation of China(51971094)
National Natural Science Foundation of China(21476088)
National Natural Science Foundation of China(21776104)
Guangdong Provincial Department of Science and Technology(2015A030312007)
PDF
