%45%75%6e%41%65%20%43%68%6f
-
Designing the Heart of Hydrogen Cars with AI... Development of Next-Generation Super Catalyst
<(From left) KAIST Ph.D. Candidate HyunWoo Chang, Professor EunAe Cho. (Top, from left) Seoul National University Professor Won Bo Lee, Dr. Jae Hyun Ryu.>
In the era of climate crisis, hydrogen vehicles are emerging as an alternative for eco-friendly mobility. However, the fuel cell, known as the ‘heart of the hydrogen car,’ still faces limitations of high cost and short lifespan. The core cause is the platinum catalyst. While it is a decisive material for generating electricity, the reaction is slow, performance degrades over time, and manufacturing costs are high. Korean researchers have presented a clue to solving this difficult problem.
KAIST announced on February 26th that the research team led by Professor EunAe Cho of the Department of Materials Science and Engineering, together with the team of Professor Won Bo Lee of the School of Chemical and Biological Engineering at Seoul National University, has developed a technology that predicts the ‘atomic arrangement’ tendency of catalysts using artificial intelligence (AI).
This technology is akin to calculating beforehand which combination is advantageous for completing a puzzle before putting it together. By having AI calculate the arrangement speed of metal atoms first, it has become possible to efficiently design catalysts with better performance. The core of this research is that ‘AI revealed the fact that zinc plays a decisive role in the platinum-cobalt atomic arrangement.’
<Schematic diagram of AI-based atomic alignment prediction>
Despite the high performance of existing platinum-cobalt (Pt-Co) alloy catalysts, very high-temperature heat treatment was required to create the ‘intermetallic (L1₀)’ structure, where atoms are regularly arranged. In this process, particles would clump together, or the structure would become unstable, posing limitations for actual fuel cell application.
To solve this problem, the research team introduced machine learning-based quantum chemistry simulations. Through AI, they precisely predicted how atoms move and arrange themselves inside the catalyst.
As a result, they discovered that zinc (Zn) acts as a mediating element that promotes atomic arrangement. The principle is that when zinc is introduced, atoms find their places more easily, forming a more sophisticated and stable structure. In other words, AI has found the ‘optimal path for atomic arrangement creation’ in advance.
< Synthesis process of Zinc-introduced Platinum-Cobalt catalyst>
The zinc-platinum-cobalt catalyst, synthesized based on AI predictions, secured both higher activity and superior long-term durability compared to commercial platinum catalysts. This is a case proving that the ‘virtual blueprint’ calculated by artificial intelligence can be implemented as a high-performance catalyst in an actual laboratory.
In particular, this technology is expected to contribute to extending catalyst lifespan and reducing manufacturing costs across core carbon-neutral industries, such as hydrogen passenger cars, hydrogen trucks requiring long-distance operation, hydrogen ships, and energy storage systems (ESS).
< Conceptual diagram of AI-based catalyst development (AI-generated image) >
Professor EunAe Cho stated, “This research is a case of utilizing machine learning to predict the atomic arrangement tendency of catalysts in advance and implementing this through actual synthesis,” and added, “AI-based material design will become a new paradigm for the development of next-generation fuel cell catalysts.”
Ph.D. Candidate HyunWoo Chang from KAIST’s Department of Materials Science and Engineering and Dr. Jae Hyun Ryu from Seoul National University’s School of Chemical and Biological Engineering participated as co-first authors in this research. The research results were published on January 15, 2026, in ‘Advanced Energy Materials,’ a world-renowned academic journal in the energy materials field. ※ Paper Title: Machine Learning-Guided Design of L1₀-PtCo Intermetallic Catalysts: Zn-Mediated Atomic Ordering, DOI: https://doi.org/10.1002/aenm.202505211
This research was conducted with the support of the National Research Foundation of Korea’s Nano & Material Technology Development Program and the Korea Institute of Energy Technology Evaluation and Planning’s Energy Innovation Research Center for Fuel Cell Technology.
2026.02.27 View 1665 -
KAIST Transforms Hydrogen Energy by Flattening Granular Catalysts into Paper-Thin Sheets
<(From Left) Ph.D candidate HyunWoo J Yang, Ph.D candidate SangJae Lee, Professor EunAe Cho, Ph.D candidate DongWon Shin>
Catalysts are the “invisible engines” of hydrogen energy, governing both hydrogen production and electricity generation. Conventional catalysts are typically fabricated in granular particle form, which is easy to synthesize but suffers from inefficient use of precious metals and limited durability. KAIST researchers have introduced a paper-thin sheet architecture in place of granules, demonstrating that a structural innovation—rather than new materials—can simultaneously reduce precious-metal usage while enhancing both hydrogen production and fuel-cell performance.
KAIST (President Kwang Hyung Lee) announced on the 21st of January that a research team led by Professor EunAe Cho of the Department of Materials Science and Engineering has developed a new catalyst architecture that dramatically reduces the amount of expensive precious metals required while simultaneously improving hydrogen production and fuel-cell performance.
The core of this research lies in the application of ultrathin nanosheet structures, with thicknesses tens of thousands of times thinner than a human hair, enabling the team to overcome both efficiency and durability limitations of conventional catalysts.
Water electrolyzers and fuel cells are key technologies for hydrogen energy production and utilization. However, their commercialization has been severely constrained by the scarcity and high cost of iridium (Ir) and platinum (Pt), which are commonly used as catalysts. In conventional particle-based catalysts, only a limited surface area participates in reactions, and long-term operation inevitably leads to performance degradation.
To address this, the research team transformed agglomerated catalyst particles into paper-like, ultrathin and laterally extended sheets. For water electrolysis, they developed ultrathin iridium nanosheets with lateral size of 1–3 micrometers and thicknesses below 2 nanometers. This structure dramatically increased the active surface area participating in reactions, enabling significantly higher hydrogen production with the same amount of iridium.
< Ultrafine Iridium Nanosheet (AI-generated image) >
In addition, the team discovered that these ultrathin nanosheets naturally formed interconnected conductive pathways on titanium oxide (TiO2), a material previously considered unsuitable as a catalyst support due to its poor electrical conductivity. As a result, titanium oxide could be stably used as a catalyst support, further enhancing durability.
The resulting catalyst achieved a 38% higher hydrogen production rate than commercial catalysts and operated stably for over 1,000 hours under high-load, industry-relevant conditions (1 A/cm2*). Notably, even with approximately 65% less iridium, the catalyst delivered performance comparable to commercial benchmarks, demonstrating a major reduction in precious-metal usage.
*1 A/cm2: a high-current condition corresponding to intensive operation of practical hydrogen-production systems
The team further applied the ultrathin nanosheet design strategy to fuel-cell catalysts, producing platinum–copper nanosheets with thicknesses again tens of thousands of times thinner than a human hair.
In fuel-cell evaluations, this catalyst exhibited a 13-fold improvement in mass activity per unit platinum compared with commercial catalysts, and delivered approximately 2.3 times higher performance in full fuel-cell tests. Even after 50,000 accelerated durability cycles, the catalyst retained about 65% of its initial performance, significantly outperforming conventional catalysts. Importantly, the same performance was achieved while reducing platinum usage by approximately 60%.
Professor EunAe Cho emphasized, “This study presents a new catalyst architecture that simultaneously enhances hydrogen production and fuel-cell performance while using far less expensive precious metals,” adding, “It represents a critical turning point for lowering the cost of hydrogen energy and accelerating its commercialization.”
<Schematic illustration of ultrathin nanosheet synthesis and transmission electron microscopy (TEM) images of the fabricated catalyst>
<Fabrication process of an ultrathin nanosheet catalyst and transmission electron microscopy (TEM) images of the fabricated catalyst>
The results of this work were published in two separate papers, both based on the shared core technology of ultrathin nanosheet architectures—one focused on hydrogen-production catalysts and the other on fuel-cell catalysts.
The iridium nanosheet study, with doctoral candidate Dongwon Shin as first author, was published online on December 10, 2025, in ACS Nano (IF 16.0).
※ Paper title: “Ultrathin Iridium Nanosheets on Titanium Oxide for High-Efficiency and Durable Proton Exchange Membrane Water Electrolysis,” DOI: 10.1021/acsnano.5c15659
The platinum–copper nanosheet study, with SangJae Lee and doctoral candidate HyunWoo Yang as co–first authors, was published online on December 11, 2025, in Nano Letters (IF 9.6).
※ Paper title: “Ultrathin PtCu Nanosheets: A New Frontier in Highly Efficient and Durable Catalysts for the Oxygen Reduction Reaction,” DOI: 10.1021/acs.nanolett.5c04848
This research was supported by the Energy Human Resource Development Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) under the Ministry of Trade, Industry and Energy, and by the Nano- and Materials-Technology Development Program of the National Research Foundation of Korea under the Ministry of Science and ICT.
2026.01.23 View 1379