Researchers at the University of Toronto have created this stunning material by merging machine learning (AI) with nanoscale engineering. The breakthrough could transform industries from aerospace to automotive.

The Quest for Stronger, Lighter Materials

For decades, engineers have sought materials that are both lightweight and strong. The goal is simple: reduce weight without sacrificing durability. This is especially critical in industries like aerospace, where every gram saved can lead to significant fuel savings and increased performance. Traditional materials like aluminum and titanium have their limits. And while carbon fiber has been a game-changer, it’s not without its drawbacks either.

In their quest to develop and push the limits of materials science, the researchers in Canada turned to nanoarchitected materials —structures designed at the nanoscale to maximize strength and minimize weight. These materials take inspiration from nature, mimicking the structures found in bones, shells, and even honeycombs. But designing these structures is no easy task. The challenge lies in creating geometries that distribute stress evenly, avoiding weak points where failure can begin.

To overcome these hurdles, researchers turned to Bayesian optimization, a form of machine learning that excels at finding the best possible design among countless options. By feeding the algorithm data from thousands of simulations, the team was able to identify the most efficient shapes for their carbon nanolattices.

“Nano-architected materials combine high-performance shapes, like making a bridge out of triangles, at nanoscale sizes, which takes advantage of the ‘smaller is stronger’ effect, to achieve some of the highest strength-to-weight and stiffness-to-weight ratios, of any material,” says Peter Serles, the first author of the new paper.

“However, the standard lattice shapes and geometries used tend to have sharp intersections and corners, which leads to the problem of stress concentrations. This results in early local failure and breakage of the materials, limiting their overall potential.

“As I thought about this challenge, I realized that it is a perfect problem for machine learning to tackle.”

The process began with the algorithm generating thousands of potential designs. Each design was tested in a virtual environment using finite element analysis, a computational method that predicts how materials will behave under stress. The algorithm then refined its designs, iterating toward structures that maximized strength and stiffness while minimizing weight.

Once the machine supplied a short list of optimized designs, the team physically created the proposed materials using two-photon polymerization, a form of 3D printing that can create structures with nanoscale precision. Using this technique, they produced lattices composed of beams just 300 to 600 nanometers thick. These lattices (6.3 x 6.3 x 3.8 millimeters), composed of 18.75 million individual cells, were then subjected to pyrolysis, a process that converts the polymer into a glassy carbon by heating it to 900 degrees Celsius in a nitrogen-rich environment.

The optimized nanolattices more than doubled the strength of previous designs. They withstood stress of 2.03 megapascals per cubic meter per kilogram of density. To put that in perspective, this is more than ten times stronger than many lightweight materials like aluminum alloys or even some forms of carbon fiber. This performance is about five times higher than that of titanium.

“This is the first time machine learning has been applied to optimize nano-architected materials, and we were shocked by the improvements,” says Serles. “It didn’t just replicate successful geometries from the training data; it learned from what changes to the shapes worked and what didn’t, enabling it to predict entirely new lattice geometries.

Strength Ironically Increases With Smaller Size

What makes these nanolattices so strong? The answer lies in the unique properties of carbon at the nanoscale. The researchers discovered that reducing the diameter of the carbon beams to just 300 nanometers led to a significant increase in strength. This is due to a phenomenon known as the “size effect,” where materials behave differently at extremely small scales.

At the nanoscale, carbon atoms arrange themselves in a way that maximizes strength. The researchers found that the outer layers of the carbon beams were composed of 94% sp²-bonded carbon, a form of carbon known for its exceptional strength and stiffness. This high-purity carbon shell, combined with the optimized geometry of the beams, allows the material to withstand immense forces without breaking.

The potential impact reaches far beyond the lab. Ultra-lightweight components may soon power planes, helicopters, and spacecraft. Lighter parts can reduce fuel demands and lower emissions.

“For example, if you were to replace components made of titanium on a plane with this material, you would be looking at fuel savings of 80 liters per year for every kilogram of material you replace,” adds Serles.

Looking ahead, the researchers plan to scale up their designs. “Our next steps will focus on further improving the scale-up of these material designs to enable cost-effective macroscale components,” adds Filleter. “In addition, we will continue to explore new designs that push the material architectures to even lower density while maintaining high strength and stiffness.”

The findings appeared in the journal Advanced Materials.