Collaborative Research: MFS-SPEED: Predictive Discovery of Sustainable Biopolymers via Multi-Attribute Descriptor System, Robotics/Machine Learning Workflow, and Open-Data Platform

About This Grant

With this award, Professors Chen, Dutta, Li, and Curtzwiler of the University of Maryland and Iowa State University are studying how to leverage artificial intelligence to accelerate the discovery of high-performance and biodegradable polymer nanocomposites with tunable properties. The team will develop an integrated platform that combines robotic platforms, artificial intelligence, and materials chemistry to accelerate the process to identify, design, and test polymer composite films with customizable properties, such as mechanical strength, optical clarity, and moisture absorption. A public, open-access database and user-friendly interface will support broad engagement across scientific, industrial, and policy communities. In parallel, the project will foster workforce development through K-12 research internships, undergraduate mentorship, and the integration of findings into university curricula. These efforts will aim to cultivate the next generation of scientists and engineers equipped to lead innovation in artificial intelligence-accelerated materials discovery. With this award, Professors Chen, Dutta, Li, and Curtzwiler of the University of Maryland and Iowa State University are studying how to integrate high-throughput robotic experimentation, explainable machine learning, and multiscale simulations to enable predictive design of biopolymer nanocomposites. The project will develop a multi-attribute descriptor framework to encode molecular structure, processing conditions, and life cycle assessment metrics for multiple biopolymer components that are generally recognized as safe. These descriptors will be used to generate and analyze thousands of composite formulations via a robotics-enabled workflow. Data from optical, mechanical, and dielectric characterization will train an ensemble of neural network models capable of accurately predicting properties. The project will apply counterfactual explanation algorithms to identify key formulation and processing features that drive high performance and support inverse design. Complementary molecular dynamics and density functional theory simulations will provide atomistic insight into the effects of ion binding and chemical modifications. The data, tools, and models from this project will be disseminated through a cloud-based platform that enables forward and reverse materials design. This framework will expand the accessible design space for biodegradable polymers and accelerate the development of next-generation materials that combine high performance with low impact on resources. 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.