For years, scientists have been hunting for futuristic materials that could outperform steel and iron — ideally, at a fraction of the environmental cost. Globally, an estimated 1,881 million tonnes of steel are consumed annually, but the traditional production of these metals is water- and energy-intensive. The steel industry alone accounts for a staggering 7% of global greenhouse gas emissions.
Enter nano-architected materials — specially engineered 3D-printed structures that combine lightweight design with remarkable strength, electrical, and thermal properties. Built from repeating units as small as 1 to 100 nanometers, these materials form intricate frameworks known as nanolattices.
Now, researchers at the University of Toronto have taken a big leap in this field. By combining machine learning with 3D nano-printing, they’ve created the strongest nano-architected material to date — with the strength of steel and the weight of Styrofoam.
Creating the strongest nano-architected material
One major weakness of conventional nanolattices has been their geometry. Shapes like triangles and squares often concentrate stress at corners and joints, making them prone to breaking — especially when scaled up.
“The majority of nanolattices use standard geometries like triangles, but these tend to break at the connection points because of stress concentrations. But we knew that this would be an ideal problem for machine learning – where it could try tons of different geometries and figure out how to avoid these failure points by reshaping the material,” Peter Serles, lead researcher, and a PhD student from the University of Toronto, told ZME Science.
For their purpose, researchers focused on nanolattices made of carbon units. They used a multi-objective Bayesian optimization machine learning algorithm, which specializes in comparing multiple outcomes and finding the best solution for a problem.
With this algorithm, the study authors ran simulations of different geometries to identify which shape provided the most balanced stress distribution and the best strength-to-weight ratio. This process ultimately led to an optimized nanolattice design with the strength of steel at the weight of Styrofoam.
They produced 18.75 million such carbon nanolattices using a 3D nano printer. This material can withstand 2.03 megapascals of stress per cubic meter per kilogram—roughly the same pressure you’d feel if a 2,000 kg car were balanced on a single soda can.
“This is the first time machine learning has been applied to optimize nano-architected materials, and we were shocked by the improvements. 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,” Serles said.
The big change a tiny nanolattice can bring
The researchers claim they can currently produce several cubic millimeters of the material. Although this isn’t at a commercial scale yet, the printing speeds are getting 100-1000x faster every few years. This suggests we may be just a few years away from using these 3D-printed nanolattices to build full-size structures.
Printing these optimized nanolattices is not only less energy and resource-intensive than traditional iron and steel production but could also help reduce fossil fuel demand and carbon emissions in various indirect ways.
For instance, “these materials have amazing potential in aviation, where lightweight materials lead to huge fuel savings without sacrificing safety and strength. For example, replacing one kilogram of steel in an aircraft with this material would save as much as 80 liters of fuel per year,” Serles told ZME Science.
A report from the International Air Transport Association (IATA) suggests that an airplane emits 3.16 kg of CO2 for every liter of fuel it burns. This means that replacing one kilogram of steel in a plane with the nano-architected material could prevent over 250 kgs of CO2 emissions every year.
Moreover, the material could also be used for cars, helicopters, and rockets. However, it is first important to scale up the design so that the nano-architected material can be used to make large components at affordable rates. Future research will focus on achieving this goal.
“In addition, we will continue to explore new designs that push the material architectures to even lower density while maintaining high strength and stiffness,” Tobin Filleter, one of the study authors and professor at the University of Toronto, said.
The study is published in the journal Advanced Materials.
