Autonomous AI Design Unlocks a Faster Route to Custom Materials

Key points

• Researchers often know what they want a material to do — conduct electricity, withstand heat, respond to stimuli or display a specific color — but turning those goals into a working chemical formula can take months or years of trial and error.
• The process is especially challenging for polymers, whose performance can change dramatically and unpredictably with even small tweaks to their molecular building blocks or how they are combined.
• Researchers at the U.S. Department of Energy’s (DOE) Arogonne Laboratory, the University of Chicago, and Purdue University have now demonstrated a faster path: An autonomous inverse-design workflow that helps scientists go from a target property to a polymer recipe with far fewer experiments.
• Autonomous workflow combines artificial intelligence, machine learning, and robotics to rapidly create polymers with precise, customizable properties.
• The approach connects three pieces that are often separate in traditional research.