Construction of Reliable Vertical Large Language Model Systems for Sustainable Catalysis: From Evidence‐Bound Knowledge Extraction to Closed‐Loop Catalyst Discovery
ChemSusChem · 2026-11-26 · 19 (22)
DOI: 10.1002/cssc.71230 ↗At a glance
Übersichtsarbeit in ChemSusChem, published online on 2026-11-18. Written by Xinlei Tu and 10 co-authors. Institution: State Key Laboratory of Green Chemical Synthesis and Conversion, College of Chemical Engineering, Zhejiang University of Technology, Hangzhou, P. R. …. 147 references. Funded by: National Key Research and Development Program of China, National Natural Science Foundation of China-Zhejiang Joint Fund for the Integration of Industrialization and Informatization, Foundation for Innovative Research Groups of the National Natural Science Foundation of China.
Abstract (original language)
Catalysis is central to energy conversion and green chemistry, but its development is constrained by fragmented literature data, complex reaction systems, difficult mechanistic identification, and reliance on empirical trial and error. Large language models (LLMs), with capabilities in unstructured text understanding, multisource knowledge integration, reasoning, and tool invocation, have been increasingly applied to literature extraction, characterization analysis, mechanistic hypothesis generation, catalyst recommendation, condition optimization, and automated experimentation. However, catalytic tasks are highly specialized, condition‐dependent, and validation‐intensive, making general‐purpose LLMs insufficient for traceable, verifiable, safe, and controllable research. This review summarizes recent progress in applying LLMs to key catalytic tasks and discusses their reliability requirements. We further propose a catalysis‐oriented vertical LLM framework consisting of “task definition–evidence access–constraint control–model adaptation–validation evaluation–feedback updating.” By integrating evidence‐constrained knowledge extraction, constraint‐aware decision‐making, and feedback‐driven closed‐loop optimization, this framework aims to reduce low‐value trial‐and‐error experiments, unnecessary reagent consumption, energy input, and waste generation, thereby supporting reliable and sustainable catalysis research.
Subjects
Author keywords
Authors and institutions
- State Key Laboratory of Green Chemical Synthesis and Conversion, College of Chemical Engineering, Zhejiang University of Technology, Hangzhou, P. R. China.
- Wenzhou-Kean University, Wenzhou, Zhejiang, P. R. China.
ORCID profiles: Qingtao Wang, Chunshan Lu, Qunfeng Zhang
Funding
- National Key Research and Development Program of China
- National Natural Science Foundation of China-Zhejiang Joint Fund for the Integration of Industrialization and Informatization
- Foundation for Innovative Research Groups of the National Natural Science Foundation of China
- Zhejiang Province Applied Basic Research Program
Cite
APA
Tu, X., Fang, R., Guo, L., Lyu, H., Zhao, Y., Li, X., Wang, Q., Feng, F., Lu, C., Zhang, Q., & Li, X. (2026). Construction of Reliable Vertical Large Language Model Systems for Sustainable Catalysis: From Evidence‐Bound Knowledge Extraction to Closed‐Loop Catalyst Discovery. ChemSusChem, 19(22). https://doi.org/10.1002/cssc.71230
Vancouver
Tu X, Fang R, Guo L, Lyu H, Zhao Y, Li X, et al. Construction of Reliable Vertical Large Language Model Systems for Sustainable Catalysis: From Evidence‐Bound Knowledge Extraction to Closed‐Loop Catalyst Discovery. ChemSusChem. 2026;19(22). doi:10.1002/cssc.71230
BibTeX
@article{Tu2026,
author = {Tu, Xinlei and Fang, Rubo and Guo, Lixia and Lyu, Haowei and Zhao, Yashuang and Li, Xurui and Wang, Qingtao and Feng, Feng and Lu, Chunshan and Zhang, Qunfeng and Li, Xiaonian},
title = {Construction of Reliable Vertical Large Language Model Systems for Sustainable Catalysis: From Evidence‐Bound Knowledge Extraction to Closed‐Loop Catalyst Discovery},
journal = {ChemSusChem},
year = {2026},
volume = {19},
number = {22},
doi = {10.1002/cssc.71230}
}
EUVIMED