AI-Enabled Materials Design of Non-Periodic 3D Architectures With Predictable Direction-Dependent Elastic Properties

Natural porous materials have exceptional properties-for example, light weight, mechanical resilience, and multi-functionality. Efforts to imitate their properties in engineered structures have limited success. This, in part, is caused by the complexity of multi-phase materials composites and by the lack of quantified understanding of each component's role in overall hierarchy. This challenge is twofold: 1) computational. because non-periodicity and defects render constructing design guidelines between geometries and mechanical properties complex and expensive and 2) experimental. because the fabrication and characterization of complex, often hierarchical and non-periodic 3D architectures is non-trivial.

Text-to-Microstructure Generation Using Generative Deep Learning

Designing novel materials is greatly dependent on understanding the design principles, physical mechanisms, and modeling methods of material microstructures, requiring experienced designers with expertise and several rounds of trial and error. Although recent advances in deep generative networks have enabled the inverse design of material microstructures, most studies involve property-conditional generation and focus on a specific type of structure, resulting in limited generation diversity and poor human-computer interaction. In this study, a pioneering text-to-microstructure deep generative network (Txt2Microstruct-Net) is proposed that enables the generation of 3D material microstructures directly from text prompts without additional optimization procedures. The Txt2Microstruct-Net model is trained on a large microstructure-caption paired dataset that is extensible using the algorithms provided. Moreover, the model is sufficiently flexible to generate different geometric representations, such as voxels and point clouds. The model's performance is also demonstrated in the inverse design of material microstructures and metamaterials. It has promising potential for interactive microstructure design when associated with large language models and could be a user-friendly tool for material design and discovery.

Deep Learning in Mechanical Metamaterials: From Prediction and Generation to Inverse Design

Mechanical metamaterials are meticulously designed structures with exceptional mechanical properties determined by their microstructures and constituent materials. Tailoring their material and geometric distribution unlocks the potential to achieve unprecedented bulk properties and functions. However, current mechanical metamaterial design considerably relies on experienced designers' inspiration through trial and error, while investigating their mechanical properties and responses entails time-consuming mechanical testing or computationally expensive simulations. Nevertheless, recent advancements in deep learning have revolutionized the design process of mechanical metamaterials, enabling property prediction and geometry generation without prior knowledge. Furthermore, deep generative models can transform conventional forward design into inverse design. Many recent studies on the implementation of deep learning in mechanical metamaterials are highly specialized, and their pros and cons may not be immediately evident. This critical review provides a comprehensive overview of the capabilities of deep learning in property prediction, geometry generation, and inverse design of mechanical metamaterials. Additionally, this review highlights the potential of leveraging deep learning to create universally applicable datasets, intelligently designed metamaterials, and material intelligence. This article is expected to be valuable not only to researchers working on mechanical metamaterials but also those in the field of materials informatics.

Machine learning unifies flexibility and efficiency of spinodal structure generation for stochastic biomaterial design

Porous biomaterials design for bone repair is still largely limited to regular structures (e.g. rod-based lattices), due to their easy parameterization and high controllability. The capability of designing stochastic structure can redefine the boundary of our explorable structure-property space for synthesizing next-generation biomaterials. We hereby propose a convolutional neural network (CNN) approach for efficient generation and design of spinodal structure-an intriguing structure with stochastic yet interconnected, smooth, and constant pore channel conducive to bio-transport. Our CNN-based approach simultaneously possesses the tremendous flexibility of physics-based model in generating various spinodal structures (e.g. periodic, anisotropic, gradient, and arbitrarily large ones) and comparable computational efficiency to mathematical approximation model. We thus successfully design spinodal bone structures with target anisotropic elasticity via high-throughput screening, and directly generate large spinodal orthopedic implants with desired gradient porosity. This work significantly advances stochastic biomaterials development by offering an optimal solution to spinodal structure generation and design.

Optimally-Tailored Spinodal Architected Materials for Multiscale Design and Manufacturing

Spinodal architected materials with tunable anisotropy unify optimal design and manufacturing of multiscale structures. By locally varying the spinodal class, orientation, and porosity during topology optimization, a large portion of the anisotropic material space is exploited such that material is efficiently placed along principal stress trajectories at the microscale. Additionally, the bicontinuous, nonperiodic, unstructured, and stochastic nature of spinodal architected materials promotes mechanical and biological functions not explicitly considered during optimization (e.g., insensitivity to imperfections, fluid transport conduits). Furthermore, in contrast to laminated composites or periodic, structured architected materials (e.g., lattices), the functional representation of spinodal architected materials leads to multiscale, optimized designs with clear physical interpretation that can be manufactured directly, without special treatment at spinodal transitions. Physical models of the optimized, spinodal-embedded parts are manufactured using a scalable, voxel-based strategy to communicate with a masked stereolithography (m-SLA) 3D printer.

Bulk Ferroelectric Metamaterial with Enhanced Piezoelectric and Biomimetic Mechanical Properties from Additive Manufacturing

Three-dimensional (3D) ferroelectric materials are electromechanical building blocks for achieving human-machine interfacing, energy sustainability, and enhanced therapeutics. However, current natural or synthetic materials cannot offer both a high piezoelectric response and desired mechanical toughness at the same time to meet the practicality. Here, a lamellar ferroelectric metamaterial was created with a ceramic-like piezoelectric property and a bone-like fracture toughness through a low-voltage-assisted 3D printing technology. The one-step printed bulk structure, consisting of periodically intercalated soft ferroelectric and hard electrode layers, exhibited a significantly enhanced longitudinal piezoelectric charge coefficient (d33) of over 150 pC N-1, as well as a superior fracture resistance of ∼5.5 MPa·m1/2, more than three times higher than conventional piezo-ceramics. The excellent printability together with the combination of both high piezoelectric and mechanical behaviors allowed us to create a bone-like structure with tunable anisotropic piezoelectricity and bone-comparable mechanical properties, showing a potential of manufacturing practical, high-performance, and smart biological systems.

Extreme mechanical resilience of self-assembled nanolabyrinthine materials

Low-density materials with tailorable properties have attracted attention for decades, yet stiff materials that can resiliently tolerate extreme forces and deformation while being manufactured at large scales have remained a rare find. Designs inspired by nature, such as hierarchical composites and atomic lattice-mimicking architectures, have achieved optimal combinations of mechanical properties but suffer from limited mechanical tunability, limited long-term stability, and low-throughput volumes that stem from limitations in additive manufacturing techniques. Based on natural self-assembly of polymeric emulsions via spinodal decomposition, here we demonstrate a concept for the scalable fabrication of nonperiodic, shell-based ceramic materials with ultralow densities, possessing features on the order of tens of nanometers and sample volumes on the order of cubic centimeters. Guided by simulations of separation processes, we numerically show that the curvature of self-assembled shells can produce close to optimal stiffness scaling with density, and we experimentally demonstrate that a carefully chosen combination of topology, geometry, and base material results in superior mechanical resilience in the architected product. Our approach provides a pathway to harnessing self-assembly methods in the design and scalable fabrication of beyond-periodic and nonbeam-based nano-architected materials with simultaneous directional tunability, high stiffness, and unsurpassed recoverability with marginal deterioration.

Substrate curvature as a cue to guide spatiotemporal cell and tissue organization

Recent evidence clearly shows that cells respond to various physical cues in their environments, guiding many cellular processes and tissue morphogenesis, pathology, and repair. One aspect that is gaining significant traction is the role of local geometry as an extracellular cue. Elucidating how geometry affects cell and tissue behavior is, indeed, crucial to design artificial scaffolds and understand tissue growth and remodeling. Perhaps the most fundamental descriptor of local geometry is surface curvature, and a growing body of evidence confirms that surface curvature affects the spatiotemporal organization of cells and tissues. While well-defined in differential geometry, curvature remains somewhat ambiguously treated in biological studies. Here, we provide a more formal curvature framework, based on the notions of mean and Gaussian curvature, and summarize the available evidence on curvature guidance at the cell and tissue levels. We discuss the involved mechanisms, highlighting the interplay between tensile forces and substrate curvature that forms the foundation of curvature guidance. Moreover, we show that relatively simple computational models, based on some application of curvature flow, are able to capture experimental tissue growth remarkably well. Since curvature guidance principles could be leveraged for tissue regeneration, the implications for geometrical scaffold design are also discussed. Finally, perspectives on future research opportunities are provided.

Ultrahigh Energy Absorption Multifunctional Spinodal Nanoarchitectures

Nanolattices are promoted as next-generation multifunctional high-performance materials, but their mechanical response is limited to extreme strength yet brittleness, or extreme deformability but low strength and stiffness. Ideal impact protection systems require high-stress plateaus over long deformation ranges to maximize energy absorption. Here, glassy carbon nanospinodals, i.e., nanoarchitectures with spinodal shell topology, combining ultrahigh energy absorption and exceptional strength and stiffness at low weight are presented. Noncatastrophic deformation up to 80% strain, and energy absorption up to one order of magnitude higher than for other nano-, micro-, macro-architectures and solids, and state-of-the-art impact protection structures are shown. At the same time, the strength and stiffness are on par with the most advanced yet brittle nanolattices, demonstrating true multifunctionality. Finite element simulations show that optimized shell thickness-to-curvature-radius ratios suppress catastrophic failure by impeding propagation of dangerously oriented cracks. In contrast to most micro- and nano-architected materials, spinodal architectures may be easily manufacturable on an industrial scale, and may become the next generation of superior cellular materials for structural applications.