Materials By Design: Rational Modeling, Optimization and Synthesis of Heterogeneous Materials
Abstract
In this project, we will develop a new design and synthesis platform to create new biologically inspired materials with multiple functions, driven by a new artificial intelligence platform that integrates physics-based multiscale modeling and machine learning with additive manufacturing. Natural evolution and adaptation has resulted in a tremendous array of biological protein materials that are involved in critical material functions across the natural world, resembled by organs, tissues and cells and extraorganismal material. In biological materials, hierarchical structures of protein and other polymer constituents in the form of globular and fibrous molecules, fibrils and fibers provide remarkable material properties, including ability to self-heal, as well as their capacity to combine robustness, strength, adaptability, mutability and general multi-functionality of a material~s properties. Given the complex hierarchical structural feature of natural materials, it is often not clear whether and how their structures at all levels are adapted to reach a certain material function. Moreover, to determine whether a given structure is the best suited one, one cannot expect to run direct numerical simulations by using conventional modeling tools, as the computational cost would be prohibitive. However, artificial intelligence, enabled by machine learning technique, now provides a novel feasible way of solving this problem and offers an efficient approach to searching a high-dimensional parameter space for optimal solutions, driven by data generated from physics-based simulations. We will optimize the properties of several protein material platforms to realize key material properties including high strength, stiffness, elasticity, toughness or their combinations by designing their molecular structure and mesoscale assembly through computational modeling. The key advantages of our new AI driven material modeling and design include: 1) Capability to optimize the chemical composition and architecture of protein-based composite fiber and membrane based on an efficient algorithm; 2) Breakthrough capability to optimize the multiscale and multiparadigm architecture of protein-based composite membranes for high sensitivity to environmental factors, as needed in sensors, electronics and for multi-purpose material applications; 3) Ability to extract 3D printing and processing parameters to precisely achieve the structure and optimized material functions as synthesis. All codes, material designs and other products and outcomes will be shared with researchers at Naval Research Labs and other DoD laboratories, to increase the impact for translation for real-world Defense applications. This work will result in the development of a new research field identified as biomateriomics that is enabled by the application of multiscale methods to multiple materials. Lessons learned from this basic research project, in particular insight into the use of universal elements to create diverse functions through the implementation of hierarchical structures with controlled features at multiple length-scales (molecular, nano, micro, meso, to macro), may also find applications in Defense strategies, war plans and integrated systems that combine intelligence, materials and human actions. Lessons learned from this basic research project ~ and the fundamental science questions we will advance, in particular insights into the use of universal elements to create optimized material functions through the implementation of hierarchical structures ~ will contribute to ensure America~s military, scientific and technological leadership in the future, and this basic research could prove essential for many technological and scientific innovations, including the design of advanced materials of reliability and energy efficiency. To that end, we anticipate close collaborations with Naval research labs and other Defense agencies.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 13, 2019
- Source ID
- N000141912375
Entities
People
- Markus J. Buehler
Organizations
- Massachusetts Institute of Technology
- Office of Naval Research
- United States Navy