Knowledge and Data-driven Exploration of Bounds on Mechanical Properties:A Review
Bounds on the mechanical properties provide fundamental guidelines for finding materials or structures with extreme mechanical performance.However,the bounds on some important mechanical properties,such as Young's modulus and tensile strength,remain unknown,while the search for target extreme materials from infinite potential materials of element combinations across the periodic table is chal-lenging.It has long been questioned:have we approached the bounds on these mechanical properties?Is there a material that is stiffer or harder than diamond?To determine the bounds on the mechanical proper-ties and find materials or structures with extreme mechanical performance,the key is to understand and quantify the structure-property relationship.Over the past decades,many attempts and achievements have been made to model the structure-property relationship,such as empirical/semiempirical formulas,first-principles calculations,machine learning,but these approaches often suffer from limitations in terms of ac-curacy,efficiency,universality,or interpretability.With the accumulation of knowledge and data,knowl-edge and data-driven understanding and modeling of structure-property relationships have shown immense potential.Recent studies within the knowledge and data-driven framework have led to the development of powerful theories for structure-property relationships.Based on these structure-property relationships,material properties can be predicted from structures,and conversely,structures can be designed for target material properties.Consequently,the bounds on some important mechanical properties have been deter-mined,and numerous materials or structures with mechanical properties close to the theoretical bounds have been designed and fabricated.Our work provides an overview of the recent progress in these explora-tions of bounds on mechanical properties.First,we present the advances in knowledge and data-driven ap-proaches for understanding and modeling structure-property relationships.Then,we review the deter-mined bounds on mechanical properties and discovered materials or structures with extreme mechanical performance based on the knowledge and data-driven approaches.Finally,we discuss the challenges,op-portunities,and some future directions in this field.
knowledge-drivendata-drivenstructure-property relationshipbounds on mechanical propertiesextreme mechanics
NETL
NSTL
万方数据
