AI‑Driven Modelling and Degradation of Biodegradable Polymers

Plastic in natural and open environments is an increasing environmental problem, even when it comes to biodegradable plastics. In the Nordic region, the colder climate slows down degradation processes. This means that even biodegradable plastics may remain longer in nature than expected.

At the same time, it is difficult to completely avoid the use of plastics and polymers in certain sectors, such as agriculture. In such cases, biodegradable polymers are considered a potentially better alternative to persistent plastics, especially when collection and recycling after use are not feasible. However, to use these materials safely, better tools are needed to understand how they actually degrade under real Nordic conditions.

Current standard tests are often conducted in laboratories under conditions that do not always reflect the complexity of the field. This can lead to misleading assessments - either overestimating biodegradation, which risks leaving microplastics in the environment, or underestimating it, causing materials to be unnecessarily rejected. Both scenarios create uncertainty that hampers innovation and the development of new sustainable and circular solutions.

This project aims to develop new AI-based models capable of predicting how quickly and to what extent biodegradable plastics break down in open environments, particularly under Nordic climate conditions. As a case study, the polymer poly(butylene adipate terephthalate) (PBAT) will be investigated, as it demonstrates better degradation performance in natural environments compared to many other bioplastics. By combining data from scientific studies with laboratory experiments, the project will build a machine‑learning‑based model that accounts for factors such as temperature and material form. The AI model has the potential to become a cost‑effective and more accurate tool for assessing biodegradability in relevant environments.