Team awarded $2 million to develop new separation membranes inspired by biological cells

For more than 60 years, synthetic separation membranes—manufactured barriers that selectively filter out unwanted substances in industrial, biomedical, desalination, gas separation and other processes—have been made the same way, using the same chemicals. But researchers at the University of Illinois Urbana-Champaign, along with the University of Massachusetts Amherst, are looking to change the established paradigm and create new, biological-based membranes that can be made without the use of toxins.

With a $2 million award from the U.S. National Science Foundation, this collaboration with researchers from the University at Buffalo and the Air Force Research Lab takes inspiration from how human cells allow some small molecules to enter through the cell wall while filtering others out.

 “The current generation of membranes is an amazing technology, but as emerging contaminants are present in our waste streams, we need new technologies that can separate them out and keep our water safe,” said Jessica Schiffman, professor of chemical engineering at UMass Amherst and project co-investigator.

Sarah Perry (ChBE PhD 10), UMass Amherst professor of chemical engineering and principal investigator of this Designing Materials to Revolutionize and Engineer our Future (DMREF) project, said the team isn’t just looking to tweak the existing chemical process — they want to change the paradigm of how membranes are created, harnessing biology-like selectivity and the chemical versatility of synthetic membranes.

“We are hoping that we could do this entirely from water, just the way that cells do,” she said. “It’s a really interesting challenge to figure out how to do this, because all these pieces exist, but nobody’s put them together. And that’s what our team is looking to do.”

Charles Sing, professor of chemical and biomolecular engineering at the University of Illinois, said his work on computational modeling of these materials – in collaboration with Viviana Monje, an assistant professor and expert in atomistic modeling at the University of Buffalo – will help inform both soft materials characterization and applied membrane development.

“I am excited for our team,” Sing said. “We are taking on a massive challenge, but that is the thrill of doing collaborative research. We have the collective expertise to tackle a variety of interrelated questions, allowing us to make atomistic and materials-level predictions, synthesize and design these same materials at the molecular level, probe these systems using X-Rays and thermodynamic characterization, and finally to compile all of this knowledge to make new membranes. This is also an incredible opportunity for the students, who will be trained to work on interdisciplinary teams, setting them up well for their future careers.”

Cecilia Leal, professor of materials science engineering at Illinois, is an expert in how actual biomembranes work and in the tools to characterize them.

“The DMREF mechanism is ideal to support our interdisciplinary team," Leal said. "Biomembranes exemplify how structural and compositional complexity can drive selective and efficient transport—yet current materials for separation membranes fall short of replicating this sophistication. To develop materials that self-organize into 3D ultrastructures with precision, scalability, and minimal intervention, we must integrate molecular design, advanced characterization, and computational modeling, all guided by data-driven approaches.”

In a cell, the membrane is made of molecules called lipids together with specialized proteins. “We want to take advantage of the ability of lipids to arrange themselves into different structures on a molecular level. However, just as the membranes of cells require specialized proteins, we want to use carefully designed natural polymers to help stabilize the resulting membranes and tune what can or cannot pass through,” said Perry.

One of the big challenges in approaching this kind of molecular-level materials design is the number of possible permutations. “It would be impossible for us to test every single possible lipid molecule and polymer,” said Perry.

In addition to running experiments in the lab, this project will use computer simulations to test how different molecules interact. “We can then feed both that data and our experimental results into a machine learning algorithm that will help to identify new promising molecules or conditions to test,” she said.

To realize this materials design workflow, the team will be collaborating with Jeff Ethier, an expert in physics-informed ML at the Air Force Research Lab.

A related challenge is the need to establish new techniques for modeling these materials. “It is extremely challenging to model lipid organization and atomistic interactions in a single simulation,” Sing said.

Sing’s work will develop new techniques to model the behavior of charged polymers and lipids as they stick together to form highly structured materials, but will be informed by Monje who will run simulations at the atomistic level and look at the specific arrangement of these molecules at the chemical level.

“Our computational modeling approach requires collaboration, due to the need to combine multiple techniques,” Sing said. “Our team is uniquely suited to tackling this open problem.”  

Ultimately, the team aims to create a foundational platform that would enable scientists to generate a membrane specifically tuned to filter out the desired material, with a broad range of applications. Within water purification, such membranes could be used for desalination and water treatment of various pollutants.

“There are also tons of applications where we need to separate out different types of molecules when we’re trying to make things,” Perry added. These include new therapeutics or antibody treatments in which the target compound is made within a cell and needs to be separated out. For many of these processes, membrane separation could be a far more efficient and cost-effective strategy, but the right membrane might not yet exist. “If such membranes did exist, we could save huge amounts of money because the way that they’re doing it right now is very energy intensive.”

University of Illinois Affiliations

Charles Sing is a professor and Director of Graduate Studies in the Department of Chemical & Biomolecular Engineering in the College of Liberal Arts & Sciences, where he also holds the James M. and Karen S. Morris Faculty Scholar appointment.

Cecilia Leal is professor of materials science and engineering in The Grainger College of Engineering and is affiliated with the Department of Bioengineering and the Carle Illinois College of Medicine. She holds the Racheff Faculty Scholar appointment.