AI to spark new recyclable plastics design
Chris Cooper, collaborators to use AI to design new plastics that can be more easily recycled for a circular economy
Imagine a world in which all types of plastic containers, packaging, carbon fiber composite bikes and knee implants could be recycled together as a single processing stream. This is the ambitious goal of researchers at Washington University in St. Louis; the University of California, Berkeley; and the National Institute of Standards and Technology.
While less than one-third of the total waste generated in the United States is recycled today, the team is using artificial intelligence to make new materials that could enable simple recycling of mixed waste streams back to pristine monomer that can then be converted to a new material with targeted properties.
Christopher Cooper, assistant professor of energy, environmental & chemical engineering in the McKelvey School of Engineering at WashU, and Brooks Abel, assistant professor of chemistry at UC Berkeley, along with Debra Audus and Sara Orski from the National Institute of Standards and Technology, will develop AI-driven design of architecturally diverse and deconstructable (ADD) polymers with a three-year, nearly $1.4 million grant from the National Science Foundation (NSF). Their goal is to design new sustainable polymers with a diverse range of properties that can be recycled without the need for costly and inefficient separation from mixed waste streams. Together, they plan to use simple feedstocks to create new types of plastics that can be converted back from a polymer to a monomer. They will develop physics-informed AI models to aid design of these plastics to meet various product specifications with different properties. They expect that their approach will allow different types of products to be integrated into a single recycling stream and accelerate the timeline to find more uses.
The results and methods developed by this research will be publicly accessible for users to use AI to guide design of their own materials. This will allow a new generation of scientists to work at the emerging intersection of polymer materials design and AI model development and use, Cooper said.
“This work will create new synthetic strategies to control chain-end and side-chain functionality, branch type and frequency and dynamic bond incorporation for polymers produced by cationic ring-opening polymerization,” Cooper said. “At the same time, a physics-informed AI will be continuously improved through active learning approaches and then used to perform inverse design to create new ADD polymers. These ADD polymers will be designed to achieve targeted properties within specified tolerances and will be validated against industry benchmarks.”
The award was made through the NSF’s Molecular Foundations for Sustainability: Sustainable Polymers Enabled by Emerging Data Analytics (MFS-SPEED) program. Additional MFS-SPEED funding is provided by Procter & Gamble, PepsiCo, Dow, BASF and IBM.
Click on the topics below for more stories in those areas
Back to NewsFaculty in this story
