SJTU Zhiyuan College Alumni Publish Breakthrough Nature Paper on AI-Driven Catalyst Discovery

October 15, 2025 Page views: 240

On September 23, 2025, Shanghai Jiao Tong University (SJTU) Zhiyuan College chemistry graduates Zhen Zhang (Class of 2021) and Zhichu Ren (Class of 2018) published a research paper as co-first authors in the world’s leading multidisciplinary science journal Nature (Accelerated Preview) titled “A multimodal robotic platform for multi-element electrocatalyst discovery.”

The study introduces CRESt (Copilot for Real-world Experimental Scientists), a next-generation intelligent platform designed to serve as a real-world laboratory assistant. By integrating multimodal LLMs (large language models) with high-throughput automated experimentation, CRESt incorporates human expertise, literature-derived knowledge, and microscopic structural information to dramatically enhance the efficiency of catalyst discovery.

Using the CRESt platform, the research team focused on the electrocatalytic oxidation of formic acid as a model system. The team’s exploration of over 900 catalyst formulations and 3,500 electrochemical tests led to the discovery of several highly efficient catalysts. All of this was achieved within the span of just three months. Furthermore, one newly developed octonary alloy catalyst demonstrated activity 9.3 times higher than that of pure palladium, highlighting its strong potential for future applications.

The research demonstrates the feasibility of a closed-loop scientific workflow, which involves hypothesis generation, robotic experimentation, data acquisition, machine-learning analysis, and new hypothesis formulation. This opens up new, viable pathways toward the future of autonomous, high-throughput materials research laboratories.

Both Zhen Zhang (Zhiyuan Cohort 10) and Zhichu Ren (Zhiyuan Cohort 7) completed their undergraduate studies in chemistry at SJTU Zhiyuan College before pursuing PhDs in materials science at MIT (Massachusetts Institute of Technology). Currently, Zhang is conducting cutting-edge research in AI for Science, while Ren completed his PhD in 2024 and subsequently founded LineWise, a startup supported by top Silicon Valley accelerator Y Combinator. LineWise is developing AI copilots for manufacturing engineers, designed to accelerate troubleshooting and root-cause analysis while minimizing unplanned production downtime.