Ultrafiltration membranes used in pharmaceutical manufacturing and other industrial processes have long relied on separating molecules by size. Now, researchers at Cornell University have created a porous material that filters molecules by chemical composition.
Two molecules of the same size and weight but different chemical properties, such as antibodies with different molecular structures, are difficult to separate using current ultrafiltration (UF) membrane technology. but, The study was published on November 13th In Nature Communications, researchers found that mixing chemically different block copolymer micelles (tiny self-assembled polymer spheres) could be used to create membranes that can filter molecules by chemical affinity.
“This is the first practical route to create UF membranes with chemically diverse pore surfaces,” he said. Ulrich WiesnerSpencer T. Olin Professor of Materials Science and Engineering and lead author of this study. “In principle, this could be achieved with a post-manufacturing process, but it would be cost-prohibitive for industry to adopt it. This new approach has the potential to truly revolutionize ultrafiltration.”
Taking inspiration from nature, such as protein channels in cells that can distinguish between metal ions of the same size using pore wall chemistry, lead author Dr. Lily Tzar. ’24 in the Wiesner group investigated how neutral and repulsive interactions between micelles influence their self-assembly within the upper separation layer. By combining up to three different block copolymers, the researchers demonstrated how these competing interactions control where different chemical reactions appear within the pores of the film surface.
“It’s a very simple idea in principle, but in practice it’s very difficult to develop experimentally,” says Wiesner, who is also a professor in the Department of Design Technology. “In particular, it is not easy to determine where different micelle chemistries are located in the upper separation layer.”
Tsar used a scanning electron microscope to image hundreds of samples and study how different micelles are arranged. Because image processing cannot easily identify chemical properties, she used machine learning to detect subtle differences in pore patterns and identify where each micelle type appears.
Co-author Fernando A. EscobedoCornell Professors of Chemical and Biomolecular Engineering (Cornell Engineering) Samuel W. Bodman and Diane Bodman performed molecular simulations to uncover the rules governing how micelles self-assemble. This is a challenge because micelles are large in number and tend to aggregate relatively far from equilibrium.
“This required us to use very coarse-grained models and numerous calibrations to capture the time and length scales involved in the experimental process,” said Escobedo, who worked with Dr. Luis Nieves Rosado. ’25 years.
This work builds on the Wiesner group’s previous progress in self-assembly of block copolymers, which allows Terrapore Technologiesa startup company led by Dr. Rachel Dolin. ’13 uses the group’s scalable block copolymer process to produce cost-effective UF membranes to separate viruses from biopharmaceuticals. This new research paves the way for companies to use the same manufacturing process to produce membranes that can perform affinity separations based on programming the pore surface chemistry.
“Companies simply want to change the recipe, or ‘magic dust,’ that goes into the same process they’ve been using for decades to give membranes a chemically diverse pore surface,” Wiesner says. “Our method represents a paradigm shift in UF-based operations and has the potential to open entirely new avenues for the use of UF membranes.”
Beyond filtration, this research could lead to new materials with new properties for applications such as smart coatings that react to the environment and biosensors that detect specific molecules. Wiesner’s group continues to research and develop methods to look deeper into the top separation layer of these materials to see how chemical patterns extend below the surface.
This research was supported by the National Science Foundation and made possible by the Cornell Center for Materials Research, Science and Engineering and the Cornell Nuclear Magnetic Resonance Facility.
Syl Kacapyr is an associate director of marketing and communications at Cornell Engineering.
