Strain learning in protein-based mechanical metamaterials.

Mechanical deformation of polymer networks causes molecular-level motion and bond scission that ultimately lead to material failure. Mitigating this strain-induced loss in mechanical integrity is a significant challenge, especially in the development of active and shape-memory materials. We report the additive manufacturing of mechanical metamaterials made with a protein-based polymer that undergo a unique stiffening and strengthening behavior after shape recovery cycles. We utilize a bovine serum albumin-based polymer and show that cyclic tension and recovery experiments on the neat resin lead to a ~60% increase in the strength and stiffness of the material. This is attributed to the release of stored length in the protein mechanophores during plastic deformation that is preserved after the recovery cycle, thereby leading to a "strain learning" behavior. We perform compression experiments on three-dimensionally printed lattice metamaterials made from this protein-based polymer and find that, in certain lattices, the strain learning effect is not only preserved but amplified, causing up to a 2.5× increase in the stiffness of the recovered metamaterial. These protein-polymer strain learning metamaterials offer a unique platform for materials that can autonomously remodel after being deformed, mimicking the remodeling processes that occur in natural materials.
Competing Interests: Competing interests statement:The authors declare no competing interest.