Low-melting point metals facilitate synthesis of Pt-based intermetallic nanocrystals

Graphical abstract

The extensive utilization of proton exchange membrane fuel cells (PEMFCs) strongly depends on stable oxygen reduction reaction (ORR) catalysts operating under harsh conditions, i.e., 0.6–1.5 V, an acidic environment with pH < 1, and 60–80 °C. Generally, structurally ordered L1₀-PtM (M being Fe, Co, Ni, etc.) intermetallic nanocrystals (i-NCs), stemming from a lower formation energy and higher cohesive energy, possess an enhanced durability, relative to the counterparts with a low ordering degree, thereby emerging as a preferred choice in the PEMFC industry in recent years. Nevertheless, the robust bonding between Pt-Pt and Pt-M, along with the high melting points of Pt and M, renders the diffusion activation barrier (E_a) elevated. Conventionally, high-temperature annealing (typically exceeding 600 °C) is necessitated to achieve atomic diffusion and ordering, which normally causes the sintering of nanoparticles, diminishes the driving force for disorder-to-order phase transitions and restricts the ordering degree [1]. Additionally, from the kinetic standpoint, the high activation barrier leads to a sluggish phase transition process, further diminishing the degree of order, often below 50%. This diminished orderliness potentially hinders the stable operational performance of fuel cells.

Very recently, Liang et al. [2] creatively have proposed an effective strategy that involves employment of low-melting-point metals (M’), such as Sn (231.9 °C), In (156.6 °C), and Ga (29.8 °C), to weaken bond strength and thereby reduce E_a, successfully lowering the ordering temperature of PtM alloy to below 450 °C and enabling the production of L1₀-Pt-M i-NCs with a high Pt content (> 40 wt.%). The latter holds significant importance for the cathode catalyst of the membrane electrode within fuel cell designed for heavy-duty vehicles, being capable of reducing the thickness of the catalyst layer and thus lowering the oxygen transmission resistance [3,4]. Taking Pt₅₀Ni₃₅Sn₁₅ as an example, Liang et al. have discovered that incorporating 15% Sn enables the kinetic ordering activation energy to drastically decline from 267.7 to 181.2 kJ/mol, allowing for attaining the ordered alloy structure at approximately 410 °C. The X-ray spectroscopy and in situ scanning transmission electron microscopy results, coupled with density functional theory (DFT) calculations, elucidate that the M’ doping causes a greater localization of electrons around M’ atoms (Fig.1(a)), weakening bond strength and thereby reducing E_a, followed by the emergence and diffusion of low-coordination surface free atoms. The latter triggers nucleation of the L1₀ phase on the (110) surface under low-temperature conditions. Then, the uninterrupted growth of the L1₀ phase, along with lattice compression, ultimately offers a highly ordered structure, as depicted in Fig.1(b). Noteworthily, beyond wet chemical synthesis, this strategy remains equally effective for alternative synthesis methods, e.g., the impregnation-annealing method, thereby highlighting its broad applicability. Moreover, the energy consumption and production time are greatly reduced by respectively 64%–81% and 50%–63% in comparison to methods employed in representative studies.

The authors of the paper conducted the performance evaluation of the membrane electrode assembled with the as-prepared catalysts subjected to acid leaching and post-annealing. Excitingly, the key performance data, i.e., power density and durability, have surpassed the 2025 targets set by the US Department of Energy (US DOE). Notably, L1₀-Pt₅₀Ni₃₅Ga₁₅/C achieved a remarkable current density of 1.12 A/cm² and a peak power density of 1.1 W/cm² at 0.7 V (Fig.1(c) and Fig.1(d)), superior to the currently reported top-tier intermetallic catalysts like PtCo i-NPs (1.08 W/cm²) [5] and Pt₃Co/Co-N-C (1.05 W/cm²) [6]. Operating under critical heat rejection limit conditions (0.67 V, 94 °C, 250 kPa_abs), L1₀-Pt₅₀Ni₃₅Ga₁₅/C delivered a rated power density of 1.15 W/cm² and a specific power density of 7.7 kW/g_Pt, exceeding the 1 W/cm² of target set by US DOE, highlighting its immense potential for practical fuel cell applications. After enduring 30000 voltage cycles, the MA loss for L1₀-Pt₅₀Ni₃₅M’₁₅/C and L1₀-Pt₅₀Co₃₅Ga₁₅/C catalysts remain below 30% (Fig.1(e) and Fig.1(f)), remarkably outperforming commercial Pt₃Co/C and Pt/C. In addition, the initial MA and stability of these catalysts exceed the US DOE 2025 targets of MA (0.44 A/mg) and MA loss (< 40% after 30000 cycles), surpassing most electrocatalysts reported previously [6–9]. Notably, the current density loss for L1₀-Pt₅₀Ni₃₅Ga₁₅/C at 0.7 V is merely 13.3% (Fig.1(e) and 1(f)). At a current density of 0.8 A/cm², both L1₀-Pt₅₀Ni₃₅M’₁₅/C and L1₀-Pt₅₀Co₃₅Ga₁₅/C catalysts exhibit voltage losses of less than 20 mV, surpassing Pt/C (86 mV) and Pt₃Co/C (50 mV), also exceeding the 2025 target (< 30 mV) set by the US DOE. Impressively, after the extended 90000 cycles under HDV conditions, the 40% L1₀-Pt₅₀Ni₃₅Ga₁₅/C electrode still maintains a high current density of 1.33 A/cm² at 0.7 V, corresponding to only 20.0% loss and surpassing the US DOE target of 1.07 A/cm² (Fig.1(g)−Fig.1(h)).

In summary, Liang et al. have employed low-melting-point metal, i.e., Sn, In, and Ga, doping to effectively lower the ordering temperature, presenting an inspiring strategy for synthesizing Pt-based intermetallic nanoalloys at comparatively lower temperatures. More importantly, this strategy is compatible with facile synthetic methods, including wet chemistry and impregnation-annealing, without the necessity for special equipment, and enables a ten-gram-scale production of intermetallic Pt-based catalysts. Notably, notwithstanding the ten-gram-scale synthesis achieved, it still falls far behind the kilogram-scale production per batch of commercial Pt/C catalyst, necessitating further improvement in synthesis efficiency for practical application. Moreover, numerous studies have emphasized the critical role of nanocrystal morphology in determining catalytic performance. However, the impact of low-melting-point metals like Sn and Ga utilized in this strategy, on the formation of i-NCs with specific morphologies, except for nanoparticles, remains unclear. As such, the feasibility of this strategy for synthesizing i-NCs with specific morphologies deserves further exploration in future.

References

[1] Zhou M, Li C, Fang J. Noble-metal based random alloy and intermetallic nanocrystals: Syntheses and applications. Chemical Reviews, 2021, 121(2): 736–795

[2] LiangJWanYLvH, . Metal bond strength regulation enables large-scale synthesis of intermetallic nanocrystals for practical fuel cells. Nature Materials, 2024,

[3] Zeng Y, Liang J, Li C. . Regulating catalytic properties and thermal stability of Pt and PtCo intermetallic fuel-cell catalysts via strong coupling effects between single-metal site-rich carbon and Pt. Journal of the American Chemical Society, 2023, 145(32): 17643–17655

[4] Tang M, Zhang S, Chen S. Pt utilization in proton exchange membrane fuel cells: Structure impacting factors and mechanistic insights. Chemical Society Reviews, 2022, 51(4): 1529–1546

[5] Yang C, Wang L, Yin P. . Sulfur-anchoring synthesis of platinum intermetallic nanoparticle catalysts for fuel cells. Science, 2021, 374(6566): 459–464

[6] Chong L, Wen J, Kubal J. . Ultralow-loading platinum-cobalt fuel cell catalysts derived from imidazolate frameworks. Science, 2018, 362(6420): 1276–1281

[7] Cheng Q, Yang S, Fu C. . High-loaded sub-6 nm Pt1Co1 intermetallic compounds with high-efficient performance expression in PEMFCs. Energy & Environmental Science, 2022, 15(1): 278–286

[8] Li J, Sharma S, Liu X. . Hard-magnet L1₀-CoPt nanoparticles advance fuel cell catalysis. Joule, 2019, 3(1): 124–135

[9] Zhao Z, Hossain M, Xu C. . Tailoring a three-phase microenvironment for high-performance oxygen reduction reaction in proton exchange membrane fuel cells. Matter, 2020, 3(5): 1774–1790

RIGHTS & PERMISSIONS

Higher Education Press 2024