Every time you drive smoothly from point A to point B, you not only enjoy the comfort of your car, but also the sophisticated engineering that makes it safe and reliable. Beyond its comfort and its protective characteristics lies a lesser known but crucial aspect: the cleverly optimized mechanical performance of microstructured materials. These essential but often overlooked materials strengthen your vehicle, ensuring durability and resistance on every journey.
Luckily, scientists at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) have thought of that for you. A team of researchers went beyond traditional trial-and-error methods to create materials with extraordinary performance through computational design. Their new system integrates physical experiments, physics-based simulations, and neural networks to overcome the gaps often seen between theoretical models and practical results. One of the most striking results: the discovery of microstructured composites – used in everything from cars to planes – that are much stronger and more durable, with an optimal balance between stiffness and toughness.
“Composite design and manufacturing is fundamental to engineering. We hope that the implications of our work will extend well beyond the field of solid mechanics. Our methodology provides a computational design model that can be adapted to various fields such as polymer chemistry, fluid dynamics, meteorology, and even robotics,” explains Beichen Li, an MIT doctoral student in electrical and computer engineering, affiliated with the CSAIL, and principal investigator of the project.
An open access article on the work was Published in Scientists progress earlier this month.
In the dynamic world of materials science, atoms and molecules are like tiny architects, constantly collaborating to build the future of everything. Nevertheless, each element must find its ideal partner and, in this case, the emphasis has been placed on finding a balance between two essential material properties: rigidity and toughness. Their method involved a large design space of two basic material types – one hard and brittle, the other soft and ductile – to explore various spatial arrangements to discover optimal microstructures.
A key innovation in their approach was the use of neural networks as surrogate models for simulations, thereby reducing the time and resources required for materials design. “This evolutionary algorithm, accelerated by neural networks, guides our exploration, allowing us to efficiently find the best performing samples,” explains Li.
Magical microstructures
The research team began their process by making 3D-printed photopolymers, about the size of a smartphone but thinner, and adding a small notch and triangular cut to each. After specialist ultraviolet light treatment, the samples were evaluated using a standard testing machine – the Instron 5984 – for tensile testing to assess strength and flexibility.
At the same time, the study combined physical tests with sophisticated simulations. Using a high-performance computing framework, the team was able to predict and refine the characteristics of materials before they were even created. The greatest feat, they say, lies in the nuanced technique of bonding different materials on a microscopic scale – a method involving an intricate pattern of tiny droplets fusing together rigid and flexible substances, striking the right balance between strength and flexibility. The simulations closely matched the physical test results, validating the overall effectiveness.
The system was complemented by their “neural network-accelerated multi-objective optimization” (NMO) algorithm, allowing navigation through the complex design landscape of microstructures, unveiling configurations with near-optimal mechanical attributes. The workflow functions as a self-correcting mechanism, continually refining predictions to bring them more in line with reality.
However, the journey was not without its challenges. Li highlights the challenges of maintaining consistency in 3D printing and integrating predictions, simulations, and real-world experiments into an efficient pipeline.
As for next steps, the team is working to make the process more usable and scalable. Li foresees a future where laboratories are fully automated, minimizing human supervision and maximizing efficiency. “Our goal is to see everything from manufacturing to testing and calculations automated in an integrated laboratory setup,” concludes Li.
Lead author and MIT professor Wojciech Matusik, along with Pohang University of Science and Technology associate professor Tae-Hyun Oh, and MIT CSAIL affiliate Bolei Deng, former postdoctoral fellow and now assistant professor at Georgia Tech , join Li in the article; Wan Shou, former postdoctoral fellow and now assistant professor at the University of Arkansas; Yuanming Hu MS '18 PhD '21; Yiyue Luo MS ’20; and Liang Shi, an MIT graduate student in electrical and computer engineering. The group's research was funded in part by Baden Aniline and Soda Factory (BASF).