Researchers at Washington University Use Artificial Intelligence for Chemistry and Material Discovery

Researchers Zhiling Zheng and Christopher Cooper at Washington University are utilizing machine learning, large language models, and data curation methods to advance chemical synthesis processes, accelerate material discovery, and create autonomous laboratories.

The Role of Artificial Intelligence in Chemistry

While machine learning does not replace human intelligence, it goes beyond human endurance to offer an essential assistive tool in the fields of chemistry and material discovery. While AI yields strong results in predicting new structures, scientists in the field are focused on physically producing these materials.

Metal-Organic Frameworks and New Materials

Zhiling Zheng works on metal-organic frameworks, which are formed by the combination of metal ions and organic building blocks and possess millions of variations. Trained on a literature-based dataset encompassing approximately 4,000 linker transformations, large language models enabled AI agents to filter options based on chemical constraints.

Polymer Synthesis and Data Curation

Christopher Cooper published an article examining how data pools for polymer synthesis should be curated. Data curation constitutes the first major step involving the transformation of data into a form that machine learning models can digest.

Dynamic Polymer Library and the Future

Cooper and Kathryn Miller developed an automated approach called the dynamic polymer description library for dynamic polymers featuring self-healing, 3D printing compatibility, and recyclability properties. Such tools allow human scientists to assume a managerial role while machines handle arduous processes in virtual environments.