This research introduces an intelligent agent system that combines Large Language Models with knowledge graphs to automate the planning of synthesis pathways for macromolecules.

• Leverages LLMs to extract and recognize complex chemical nomenclature
• Stores chemical data in structured knowledge graphs for improved accessibility
• Fully automates the retrosynthesis planning process for materials chemistry
• Addresses a critical challenge in polymer science where nomenclature is intricate and often non-unique

For materials engineers, this breakthrough offers a powerful tool to accelerate materials development by efficiently identifying reliable synthesis pathways—potentially reducing research time and increasing innovation in advanced materials.

Automated Retrosynthesis Planning of Macromolecules Using Large Language Models and Knowledge Graphs