Geometry‐Encoded Programmable Mechanics in Single‐Material Dual‐Phase Metamaterials

Programmable metamaterials that exhibit prescribed mechanical responses and adaptive deformation under external loading are highly desirable for multifunctional engineering applications. However, most existing designs rely on multi-material systems, which pose significant fabrication challenges with conventional additive manufacturing. Inspired by the unique soft-hard heterogeneous architecture of nacre, this study introduces a novel class of dual-phase (DP) metamaterials where spatially encoded soft and hard phases are realized through bending-dominated and stretching-dominated lattice architectures, respectively. By systematically varying the spatial coding patterns of soft-hard phases, representative DP metamaterials are shown to exhibit programmable nonlinear mechanical responses and tailored failure processes, achieved through geometry-based mechanical encoding governed by phase interactions and internal stress redistribution. Notably, the engineered sequenced failure processes and phase-coupling-induced strengthening effects lead to significantly enhanced energy absorption compared with the constituent architectures, while enabling customizable plateau stress. To efficiently explore the vast design space of DP metamaterials, a data-driven framework is then developed to model the relationship between spatial encodings and nonlinear mechanical responses. The trained model enables rapid and accurate inverse design of DP metamaterials matching the complex target responses for multifunctional applications. Overall, this work establishes a new geometry-based strategy for achieving highly programmable mechanical responses in single-material metamaterials.