From biomatter to bioplastics: A perspective on modeling, structure, and data-driven design

Biomatter-derived bioplastics—produced directly from unrefined seaweed, fungi, bacterial cellulose, and other biological feedstocks—offer a path toward manufacturing polymers with rich molecular compositions and hierarchical architectures not easily achieved synthetically. These materials inherit multiscale features from their biological origins, including embedded protein–carbohydrate–lipid networks, anisotropic structures, and chemically heterogeneous domains that influence phase behavior and mechanical performance. However, the lack of quantitative understanding linking feedstock composition, processing conditions, and resulting materials properties hinders broader design and optimization. This article outlines current challenges in characterizing and modeling these systems and highlights emerging approaches, including molecular simulations, microscopy-guided finite element methods, and physics-informed machine learning that begin to connect structure and function across scales. We emphasize the critical role of interpretable, data-efficient AI methods for inverse design, particularly in data-limited regimes.

Duke Scholars

Author L. Catherine Brinson Thomas Lord Department of Mechanical Engineering and Materia ...

Author Cynthia D. Rudin Computer Science