Voids, or empty spaces, exist within matter at all scales, from the astronomical to the microscopic. In a new study, researchers from the University of Wisconsin-Madison and the University of Illinois Urbana-Champaign used high-powered microscopy, computer simulations and mathematical theory to unveil nanoscale voids in three dimensions. This advancement is poised to improve the performance of many materials used in the home and in the chemical, energy and medical industries — particularly in the area of filtration, such as water treatment and desalination.
Magnification of common filters used in the home shows that, while they look like a solid piece of material with uniform holes, they are actually composed of millions of randomly oriented tiny voids that allow small particles to pass through. In some industrial applications, like water and solvent filtration, paper-thin membranes make up the barriers that separate fluids and particles.
“The materials science community has been aware of these randomly oriented nanoscale voids within filter membranes for a while,” says Falon Kalutantirige, a University of Illinois Urbana-Champaign graduate student. “The problem was that the complex structure of the membrane as a whole — which looks like nanoscale mountain ranges when magnified — was blocking our view of the void spaces. Because we could not see them, we couldn’t fully understand how they affected filtration properties. We knew that if we could find a way to see them, we could then figure out how they work and ultimately improve filter membrane performance.”
Ying Li
The study, directed by Illinois materials science and engineering professor Qian Chen and University of Wisconsin-Madison mechanical engineering professor Ying Li, is the first to integrate materials science and a mathematical concept called graph theory to help image and map out the random placement of these voids within filtration materials. The researchers published their findings April 11, 2024, in a paper in the journal Nature Communications.
Building on a previous study that used laboratory models, the new study focuses on far more complex membranes used in industrial applications, such as desalination plants.
“The surfaces of the membranes we studied in this work look flat to the naked eye, but when we zoomed in using transmission electron microscopy, electron tomography and atomic force microscopy, we could observe these voids nestled within these nanoscale mountainous landscapes that we call crumples,” says Kalutantirige, the study’s first author.
However, the team needed a means to measure and map these features to build a quantitative predictive model and gain a more holistic picture of the membrane surfaces.
“Mapping and measuring alone will work for materials with a regular or periodic structure, making it mathematically simple to scale up our models and predict how structural properties will influence the material’s performance,” Chen says. “But the irregularity we observed in our study pushed us to use graph theory, which gives us a mathematical way to describe this heterogeneous and messy — but practical — material.”
Graph theory helped the team finally gain a more holistic understanding of the filter membrane structure, which led them to discover a strong correlation between the unique physical and mechanical properties of random empty space and improved filtration performance.
To understand the formation of these structural features, the University of Wisconsin-Madison researchers used computer simulations to study the manufacturing process of the filter membranes. They discovered a unique oligomer coalescence-and-growth process during the membrane formation process, which leads to irregular and heterogeneous membrane structures. In turn, these structural complexities determine the performance of filter membranes, such as solvent permeance and selectivity.
“Our computer simulations provide molecular-level insights into the membrane structure as well as the transport mechanisms,” Li says. “Electron tomography can acquire the 3D microstructures of these membranes at nanometer resolution, which ties back to our computer simulations of membranes. Thus, the crosstalk among membrane manufacturing, 3D nanoscale morphology and computer simulations can serve as a new framework for the predictive design of separation materials. Such a convergent research model could enable solutions for making our food, water and energy systems more sustainable.”
The researchers say their method is a very universal technique for describing materials composed of irregular structures.
The U.S. Department of Energy, the Air Force Office of Scientific Research and the National Science Foundation supported this research.
Jinlong He from UW-Madison also contributed to the study.
A version of this story was originally published by the University of Illinois Urbana-Champaign.
Featured image caption: This graphic, titled “Beyond Nothingness,” was produced using computational modeling and portrays a highly magnified surface of a water filtration membrane as a mountainous landscape, with computational data points as the starry dark universe in the background. Graphic courtesy Falon Kalutantirige.