New AI-driven approach by UMD researchers to create biodegradable plastics
Last update on Jun 9, 2026
Finding biodegradable substitutes for single-use plastics has long been constrained by one fundamental obstacle: the combinatorial scale of the problem. With over 23 natural components – including cellulose, gelatin, carrageenan, gluten, sorbitol, and glycerol – testing every possible formulation combination by conventional trial-and-error methods would, according to UMD assistant professor Po-Yen Chen, take an estimated 1.8 million years.
A research program at the University of Maryland is now addressing that constraint directly, combining AI, robotics, and computer modeling to compress that timeline to months.
Replicate performance of conventional plastics without environmental persistence
The project, led by Chen and supported by a $2 million grant from the National Science Foundation, focuses on identifying natural substances that replicate the functional performance of conventional plastics – strength, opacity, moisture management – without the environmental persistence associated with standard polymer systems.
"It's a massive pollution problem with no single solution," said Chen. The United Nations Environment Program estimates global plastic production at more than 430 million metric tons annually, with approximately half categorized as single-use. Of that volume, around 11 million tons enters lakes, rivers, and oceans each year, with certain materials persisting in the environment for 400 years or more.
The technical challenge is compounded by uncertainty around degradation pathways. Polyvinyl alcohol (PVA), for example – the film material used in laundry pods – has an unresolved biodegradation profile following standard wastewater treatment. Separately, studies suggest that exposure to plastic additives such as BPA and phthalates may disrupt hormones and could be associated with inflammation, oxidative stress, and DNA damage, though the full scope of effects from lifetime plastic exposure remains an active area of research. One study also indicates that microplastic presence may be reducing oceans' capacity to absorb atmospheric carbon, though these findings are at an early stage.
Food packaging film extending food shelf life
Within the UMD program, AI-driven modeling operates as a bidirectional system: it predicts material characteristics – including tensile strength and adhesion – based on composition, and also generates candidate formulations matching performance parameters specified by the user. In a collaboration with the FDA, the team produced a food packaging film incorporating copper and chitosan, a carbohydrate derived from shellfish shells. The resulting material is reported to be transparent, compostable, moisture-regulating, and antimicrobial. In a comparative trial, the chitosan-based film outperformed standard plastic wrap in preserving cucumbers over 15 days and extended shelf life for avocados.
Intended applications extend beyond food packaging. Chen's lab is developing material candidates for office supplies, nonflammable battery packaging, UV-blocking packaging, and biodegradable medical delivery systems.
View 2400+ food packaging grades in the Master Catalog of Plastics.
"We don't expect to get rid of all the plastic already out there," said Chen, "but we hope to vastly reduce future production, and pollution, by creating replacements we know to be safe."
Parallel research at the University of Maryland Center for Environmental Science is using AI to accelerate microplastic identification in water and tissue samples – mapping distribution, movement through water systems, and biological exposure in species such as oysters in the Chesapeake Bay.
"What worries me most is how ubiquitous microplastics are and how little we know about what that means to ecosystems and to our health," said Christine Knauss, assistant research scientist at the center. "Developing benign alternatives is one of the most important things we can do."
You can review biodegradable grades from top suppliers and access technical data in the Master Catalog of Plastics.