Biodegradable alternatives to single-use plastics: Researchers uses AI to accelerate study
Last update on Jun 9, 2026
Pre-measured laundry pods that combine detergent, fabric softener, and brightener entered the market more than a decade ago. The pods are made of polyvinyl alcohol (PVA), and according to the researchers, it remains unclear whether PVA fully biodegrades after standard wastewater treatment, a question with potential implications for ecosystems, wildlife, and human health.
That question, alongside the broader presence of plastics in everyday products such as food wrap and bottles, prompted engineering researchers at the University of Maryland (UMD) to examine materials science for biodegradable alternatives: natural substances intended to match the performance of conventional plastic without comparable residues.
Compressing testing timeline with AI and robotics
The project is led by Po-Yen Chen, Assistant Professor of chemical and biomolecular engineering, and is supported by a $2 million grant from the National Science Foundation. It stems from a UMD Grand Challenges Grant. "It's a massive pollution problem with no single solution," Chen said.
Studies cited by researchers suggest that exposure to plastic additives such as BPA and phthalates may disrupt hormones and may contribute to inflammation, oxidative stress, and DNA damage, and that some plastics may be associated with increased cancer risk. A separate study indicates that the presence of microplastics could reduce the ocean's capacity to absorb atmospheric carbon. These findings are reported as preliminary and conditional rather than settled.
The United Nations Environment Programme estimates that more than 430 million metric tons of plastic are produced worldwide each year, the majority for packaging. By this estimate, roughly half is single use, an estimated 11 million metric tons reaches lakes, rivers, and oceans, some types persist in the environment for 400 years or more, and less than 10% is recovered for reuse.
The primary constraint on this kind of materials research has been time. Identifying a formulation that meets a product's requirements — properties such as strength, opacity, and moisture management — is difficult, and testing every combination and ratio of biodegradable substances by conventional trial and error is not feasible. According to Chen, even the team's current library of 23 natural components, including cellulose, gelatin, carrageenan, and gluten, alongside additives such as sorbitol and glycerol, would require 1.8 million years to test exhaustively.
AI, combined with robotics and computer modeling, compresses that timeline. "We can do those same tests in just a few months," Chen said, comparing the process to adjusting ingredients in a recipe until the result meets the target with robots adjusting composition and incorporating what they learn at each step. The team's models operate in two directions: they predict material characteristics such as tensile strength and stickiness from composition, and they suggest natural substitutes that match features specified by the user.
Accelerating the identification of microplastics
The team focuses on single-use plastics such as food packaging, which represent a large share of both production and waste. In a project with the FDA, the researchers developed a food film containing copper and chitosan, a carbohydrate derived from shellfish shells — that they describe as transparent, compostable, moisture-regulating, and antimicrobial. In a comparison study, the team reports that the film outperformed standard plastic wrap in preserving cucumbers, limiting mold and shriveling over 15 days, and extended the shelf life of avocados.
Chen's lab is also examining applications beyond food packaging, including office supplies such as file folders and badge holders, non-flammable battery packaging, UV-blocking packaging, and biodegradable medical delivery systems. "We don't expect to get rid of all the plastic already out there," he said, "but we hope to vastly reduce future production, and pollution, by creating replacements we know to be safe."
Microplastics remain a particular focus. Christine Knauss, an assistant research scientist at the UMD Center for Environmental Science, studies how oysters interact with and respond to microplastics in the Chesapeake Bay, and uses AI to accelerate the identification of microplastics in water and tissue samples. The work examines where microplastics are present, how they move through the water system, the likelihood of oyster exposure, and whether microplastics affect oyster growth, development, and respiration.
According to Knauss, the oysters have remained relatively resilient in the bay, while the effect of the material on human bodies remains an open question. "These pollutants are everywhere, in our soil, water, food, and both indoor and outdoor air; scientists have even found them in our lung tissue," she said. She characterized the development of alternative materials as a research priority, citing the volume of single-use plastic currently produced and entering the environment.