Collaborative Research: MFS-SPEED: D-Diameter - Data-Driven Innovation Advancing Macromolecular Engineering Towards Efficient Recycling

About This Grant

With this award, Professors Linda Broadbelt and Eugene Chen of Northwestern University (NU) and Colorado State University (CSU), respectively, are studying data-driven design of recyclable plastics using artificial intelligence and machine learning (AI/ML). A large majority of today's commodity polymers were invented in the 1930s – 1950s and further developed for performance, durability, profitability, scalability, and disposability, rather than for efficient reuse of resources at their end of life. The linear economy framework of the "mine, make, use, dispose" model not only accelerated depletion of finite natural resources but also brought about enormous material value loss to the economy. The NU-CSU team proposes to address these challenges by using AI/ML coupled with experimental approaches to design chemically recyclable polymers, materials that can be selectively and rapidly depolymerized back to their monomers for virgin-quality polymer reproduction, improving energy efficiency and use of domestic resources. The project will build on the AI/ML tools to develop new approaches and open-source software that can be applied to effectively design and realize next-generation reusable polymers in real-world applications and marketplaces. To realize the potential of AI/ML applied to the design of plastics, an informed and educated workforce is critical. The project will train multiple graduate students and postdoctoral researchers in the AI/ML approach that will be developed, and students at the undergraduate and K-12 levels will also be educated through research and educational opportunities, including disseminating software for public use. With this award, Professors Linda Broadbelt and Eugene Chen of Northwestern University (NU) and Colorado State University (CSU), respectively, are studying data-driven design of polymers using artificial intelligence and machine learning (AI/ML) coupled with experimental design of new materials. There are three main elements of the project. The first aim will address critical fundamental knowledge gaps facing data-driven polymer design for reuse by combining AI/ML-guided theory and experiment at all stages, from exploratory synthesis of bio-based monomers designed for intrinsically circular polymers through polymer characterization and performance testing to polymer end-of-life management, including closed-loop recycling and biodegradability. The second aim will establish biodegradable polyester circularity through modeling cyclic oligomers, achieve polyurethane and nylon circularity through ML-guided monomer design, and predict environmental lifetimes of polymers based on their experimental (bio)degradation kinetic profiles. Such an approach will create insightful links across traditional interfaces along the polymer manufacturing chain. Finally, the research team will develop design principles for intrinsically circular polymers with tunable and advanced properties and create new fundamental understanding of de/polymerization mechanisms via density functional theory, kinetic modeling, and AI/ML. This Molecular Foundations for Sustainability: Sustainable Polymers Enabled by Emerging Data Analytics (MFS-SPEED) award is co-funded by the NSF through the Division of Chemistry (CHE), the Directorate for Mathematical and Physical Sciences (MPS), and the Division of Innovation and Technology Ecosystems (ITE) in the Directorate for Technology, Innovation, and Partnerships (TIP). Additional MFS-SPEED funding is provided by Procter & Gamble, PepsiCo, Dow, BASF, and IBM. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.