cs.LGNov 10, 2026

Adaptive catalyst discovery emerges from composable scientific intelligence

Authors: Wentao Li, Qilong Cai, Qi Lei, Jinxing Chen, Ruixuan Chen, Jiangjie Qiu, Botian Wang, Kai Zhao, 

Organizations: Beijing Key Laboratory of Artificial Intelligence for Advanced Chemical Engineering Materials, State Key Laboratory of Chemical Engineering and Low-Carbon Technology, Department of Chemical Engineering, Tsinghua University, Beijing, China. · Department of Chemistry, National University of Singapore, Singapore, Singapore · Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China · State Key Laboratory of Advanced Optical Polymer and Manufacturing Technology, Wanhua Chemical Group Co., Ltd., Yantai 264000, P. R. China · 

Abstract

Catalysis underpins chemical manufacturing, yet knowledge for catalyst design remains dispersed across experimental observations, computational data and catalysis subdisciplines. To overcome this challenge, we present Composable Catalytic Intelligence (CatAI), a physics-grounded modular AI framework that learns 1.8 million structure-property relations and organizes them into 30 physicochemical encoders. This reusable intermediate layer enables task-specific models to guide catalyst discovery for targeted reactions and objectives. Across benchmark tasks, CatAI achieves a 7.8% lower task-averaged prediction error than the next-best models. In experimental validation, CatAI identified a new confined-passable pocket catalyst applicable in industrial homogeneous polymer synthesis with 9 times the activity of the commercial benchmark. In parallel, it guided autonomous closed-loop optimization of a highly active Cu-based geminal-atom photocatalyst for C-O coupling reactions surpassing the traditional Ni-based systems. This approach connects property-level data to catalytic objectives, supporting physicochemical knowledge reuse across mechanistically distinct systems through computational screening and iterative model-experiment feedback.

Explore similar work

CardsList
  1. CatalyticMLLM: A Graph-Text Multimodal Large Language Model for Catalytic Materials

    May 17, 2026Yanjie Li, Jian Xu, Xu-Yao Zhang +4Multimodal Large Language ModelsMaterials Property Prediction

  2. LLM-Driven Discovery of High-Entropy Catalysts via Retrieval-Augmented Generation

    Mar 16, 2026AI Scientists, Xinyi Lin, Danqing Yin +1Materials DiscoveryMaterials Science

  3. Toward Controllable Catalyst Inverse Design via Large-Scale Autoregressive Pretraining

    Jun 16, 2026Dong Hyeon Mok, Jonggeol Na, Seoin BackMaterials ScienceAutoregressive Generation