Unstructured growth of irregular architectures for optimized metastructures

Journal of the Mechanics and Physics of Solids, Год журнала: 2024, Номер 192, С. 105787 - 105787

Опубликована: Июль 17, 2024

Язык: Английский

---

Mechanical characterisation of novel aperiodic lattice structures

Materials & Design, Год журнала: 2023, Номер 229, С. 111922 - 111922

Опубликована: Апрель 12, 2023

This paper compares the mechanical properties of a class lattice metamaterials with aesthetically-pleasing patterns that are governed by mathematics aperiodic order. They built up ordered planar rod networks and exhibit higher non-crystallographic rotational symmetries. However, they lack translational symmetry associated periodic metamaterials. We present schematics illustrating their development based on pattern-unique mathematical substitution rules exploit numerical framework from previous work to demonstrate fascinating near-isotropic properties. The structures compared well-known hexagonal respect elastic anisotropy measure proximity bulk shear moduli Hashin–Shtrikman-Walpole limits. study lends insight into cost-constrained benefits introducing additional connectivities between aperiodically-ordered point sets. results show lattices have potential yield superior ones subject rigidity underlying shapes constitute pattern. inherent ‘near-isotropy’ these structures, even uniform strut thicknesses at low fractional densities, varied orientation struts them as promising mesoscale architecture for solving complex multi-axially loaded structural optimization problems, providing inspiration this study.

Язык: Английский

---

De Novo Atomistic Discovery of Disordered Mechanical Metamaterials by Machine Learning

Advanced Science, Год журнала: 2024, Номер 11(13)

Опубликована: Янв. 25, 2024

Abstract Architected materials design across orders of magnitude length scale intrigues exceptional mechanical responses nonexistent in their natural bulk state. However, the so‐termed metamaterials, when scaling bottom down to atomistic or microparticle level, remain largely unexplored and conventionally fall out coarse‐resolution, ordered‐pattern space. Here, combining high‐throughput molecular dynamics (MD) simulations machine learning (ML) strategies, some intriguing families disordered metamaterials are discovered, as fabricated by melt quenching exemplified herein lightweight‐yet‐stiff cellular featuring a theoretical limit linear stiffness–density scaling, whose structural disorder—rather than order—is key reduce exponent is simply controlled bonding interactions directionality that enable flexible tunability experimentally. Importantly, systematic navigation forcefield landscape reveals that, in‐between directional non‐directional such covalent ionic bonds, modest bond most likely promotes packing polyhedral, stretching‐dominated structures responsible for formation metamaterials. This work pioneers bottom‐down scheme formatted disorderly, unlocking untapped field leveraging disorder devising atomistically and, potentially, generic conventional upscaled designs.

Язык: Английский

---

Deep Neural Operator Enabled Concurrent Multitask Design for Multifunctional Metamaterials Under Heterogeneous Fields

Advanced Optical Materials, Год журнала: 2024, Номер 12(15)

Опубликована: Март 26, 2024

Abstract Multifunctional metamaterials (MMM) bear promise as next‐generation material platforms supporting miniaturization and customization. Despite many proof‐of‐concept demonstrations the proliferation of deep learning assisted design, grand challenges inverse design for MMM, especially those involving heterogeneous fields possibly subject to either mutual meta‐atom coupling or long‐range interactions, remain largely under‐explored. To this end, a data‐driven framework is presented, which streamlines MMMs fields. A core enabler implicit Fourier neural operator (IFNO), predicts distributed across metamaterial array, thus in general at odds with homogenization assumptions. Additionally, standard formulation problem covering broad class together gradient‐based multitask concurrent optimization identifying set Pareto‐optimal architecture‐stimulus (A‐S) pairs. multiclass blending proposed synthesize inter‐class meta‐atoms anchored on geometric motifs, while enjoying training‐free dimension reduction built‐it reconstruction. Interlocking three pillars, validated light‐by‐light programmable nanoantenna, whose involves vast space jointly spanned by quasi‐freeform supercells, maneuverable incident phase distributions, conflicting figure‐of‐merits (FoM) on‐demand localization patterns. Accommodating all challenges, can propel future advancements MMM.

Язык: Английский

---

Mechanical Performance of Copper‐Nanocluster‐Polymer Nanolattices

Advanced Materials, Год журнала: 2024, Номер 36(26)

Опубликована: Март 30, 2024

A type of copper-nanocluster-polymer composites is reported and showcased that their 3D nanolattices exhibit a superior combination high strength, toughness, deformability, resilience, damage-tolerance. Notably, the strength toughness ultralight in some cases surpass current best performers, including alumina, nickel, other ceramic or metallic lattices at low densities. Additionally, are super-resilient, crack-resistant, one-step printed under ambient condition which can be easily integrated into sophisticated microsystems as highly effective internal protectors. The findings suggest that, unlike traditional nanocomposites, laser-induced interface fraction ultrasmall Cu

Язык: Английский

---