Mining extreme properties from a large metamaterial database
Jiaxin Chen, Zhuoyi Wei, Xiaoyujie Xiao, Kai Wei, Xujing Yang, Zhaoliang Qu, Daining Fang
Abstract
Truss metamaterials exhibit a wide range of properties due to their unique node-strut architectures, which are artificially engineered through a delicate design process. However, their advanced applications are presently constrained by limited architectures and property ranges. Here, we propose a framework that systematically encodes architectural topologies and generates a comprehensive architecture-property database of over 1.8 million truss metamaterials. This database reveals numerous architectures with extreme properties, including Young’s moduli near the Voigt bound, programmable Poisson’s ratios from extremely negative to positive, and exceptional isotropic bi-mode. Moreover, we introduce the concept of mechanical isomerism. This mechanical isomerism uncovers the underlying mapping from symmetric and asymmetric architectures to extreme properties through the study of architectural variations. Our findings bridge theoretical design and engineering requirements in mining extreme properties from truss metamaterials, further enabling data-driven design, shape optimization, and advanced manufacturing. This work mines extreme mechanical properties in 2D truss metamaterials via an architecture design method and a large database. Main findings include extreme Young’s modulus, a wide range of Poisson’s ratio, extreme bi-mode, and mechanical isomerism.