Tiannan Ye's Team Publishes in Nature Communications: Machine learning-assisted Ru-N bond regulation for ammonia synthesis
发布时间:2025年08月29日

Recently, the research team led by Professor Ye Tiannan from the Frontiers Science Center for Transformative Molecules at Shanghai Jiao Tong University (SJTU) achieved a significant research breakthrough. Their findings, titled "Machine Learning-Assisted Rational Design of Intermetallic Ammonia Synthesis Catalysts", have been published in Nature Communications (Nat. Commun.).

Article abstract:

Intermetallic compounds (IMCs) with their ordered crystal structures and tunable compositions present new opportunities for developing efficient ammonia synthesis catalysts. However, the vast chemical space of IMCs makes traditional trial-and-error approaches ineffective for rapid catalyst screening. To address this challenge, Professor Ye's team developed an innovative machine learning-based workflow that combines high-throughput computational screening with experimental validation. Using this approach, the team successfully designed and synthesized a novel Sc₁/₈Nd₇/₈Ru₂ catalyst that demonstrates remarkable catalytic performance under mild conditions. The catalyst achieves a specific activity of 8.18 mmol m⁻² h⁻¹, offering a promising green alternative to the conventional Haber-Bosch process.

The study established a comprehensive "prediction-validation-mechanism" framework for materials design. By using N₂ and N adsorption energies as key descriptors, the researchers constructed an activity volcano model that identified Sc₁/₈Nd₇/₈Ru₂ as an optimal candidate. Detailed experimental and theoretical analyses revealed that Sc doping plays a dual role: it optimizes the electronic structure of Ru sites to enhance N₂ activation while simultaneously improving hydrogen poisoning resistance through the formation of lattice interstitial sites.

This work not only provides fundamental insights for developing sustainable ammonia synthesis technologies but also establishes a new paradigm for intelligent catalyst design. The successful integration of machine learning with materials science opens new avenues for accelerating the discovery of high-performance catalytic materials, with important implications for green chemical engineering applications.