Authors
Haoxiang Li, Zhiyuan Ma, Chenxi Huyan, Dong Liu, Fei Chen, Jiangxuan Song, Jun Liu, Liqun Zhang
Published in
Materials horizons. Nov 20, 2026. Epub Nov 20, 2026.
Abstract
The transition toward sustainable polymers necessitates coordinated optimization across the entire materials life cycle. Artificial intelligence (AI) and machine learning (ML) are increasingly deployed to accelerate polymer discovery, yet progress in molecular design, processing and recycling is commonly assessed within individual stages. This review critically examines the shift from single-stage property prediction toward a lifecycle-integrated AI framework. We consider five interconnected dimensions-feedstock, processing, use-phase, end-of-life and system-level impacts-and evaluate the data and model outputs that link them. Representative studies reveal how polymer structure and processing history shape recovery pathways, how manufacturing results feed back into formulation design, and how predicted reaction yields enter process balances and environmental and economic assessment. We distinguish single-stage prediction, sequential coupling, feedback-guided multi-stage optimization and optimization incorporating life cycle assessment (LCA) and techno-economic analysis (TEA), assessing the evidence for progress from laboratory demonstrations toward scalable implementation. Finally, we identify priorities in polymer data, experimental feedback, uncertainty propagation and laboratory-to-plant transfer for advancing AI-enabled circular polymer development.
PMID:
43062886
Bibliographic data and abstract were imported from PubMed on 21 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 4
- Comments 0