Bones, bread, coral reefs, concrete, ceramics, sponges, and even stars, are all made of porous materials. Found in everything from nature’s most delicate structures to humanity’s most advanced technologies, porous materials span an incredible range of forms and functions. Their tiny internal voids make them lightweight yet strong, great at absorbing shocks, filtering fluids, and insulating against heat or sound. But understanding how these materials behave and how to design better materials remains a major challenge. Their complex internal geometry and the way it evolves under stress makes them incredibly challenging to simulate across scales, even with today’s most powerful supercomputers.
A new National Science Foundation project is forging a path to that future. With it, Price Engineering’s Pania Newell will develop a hybrid computational approach that incorporates quantum algorithms for predicting the behavior of these porous materials. Determining how to best combine quantum and classical techniques in this field could significantly reduce the development time for innovations in high demand areas, including aerospace, construction, biomedicine, and others.
Newell, an associate professor in the Department of Mechanical Engineering and elected affiliate of Scientific Computing and Imaging Institute (SCI), aims to integrate cutting-edge quantum solvers with a well-established classical computing approach — finite element methods (FEM) — creating a hybrid approach to simulate and predict the behavior of porous materials. She will then use quantum topology and homogenization techniques to validate their predictions with high-resolution experimental imaging.
When studying complex systems, researchers often use FEM to model behavior at various length scales. In multiscale modeling, detailed simulations at the microscale inform the behavior of larger, macroscale systems. However, capturing fine-scale features accurately, especially when they evolve over time, can be extremely computationally expensive, even before scaling up to the full system.
“The geometry, size distribution, and connectivity of pores play a crucial role in determining the mechanical response of the materials we’re interested in,” says Newell, “but even at the microscale, simulations of porous materials can take hours to days to complete, depending on the complexity of the geometry, material heterogeneity, and boundary conditions.”
“Ultimately, our goal is to enable very fast simulations, which will help us design materials that are lighter, stronger, and more efficient,” she says.
Newell’s approach takes advantage of IBM quantum solvers, computing time on which have only recently been made available to outside researchers. The unique architecture of quantum platforms makes them particularly suited to solving a class of problems known as partial differential equations (PDEs). Newell will also use quantum-based topological techniques, to accurately, efficiently, and rapidly characterize the intricate internal structures of porous materials, bridging the microscale to macroscale.
The materials Newell will focus on are polyurethane foams; common versions of these foams are ubiquitous in packaging, but specialized formulations are sought after in a variety of high-tech applications. Integrating the results from the quantum solvers into a traditional FEM analysis, Newell will conduct experimental validation on these foams, aiming to predict the dynamics of crack formation and propagation.
“By leveraging quantum topology for capturing the complex microstructure of porous materials and quantum algorithms for solving PDEs,” Newell says, “we believe can significantly enhance both computational efficiency and the robustness of predictive modeling in porous materials.”