As climate change and population growth place mounting pressure on global water resources, communities around the world are seeking sustainable ways to reclaim water from nontraditional sources such as stormwater, agricultural runoff and municipal wastewater.
A team of researchers led by Menachem Elimelech and his former postdoctoral researcher Yanghua Duan at Rice University has taken a major step toward solving one of water purification's biggest puzzles: how to best design catalytic membranes that simultaneously filter and transform contaminants in a single step.
"Our work addresses a long-standing limitation in the field," said Elimelech, the Nancy and Clint Carlson Professor of Civil and Environmental Engineering. "Until now, most progress in reactive nanofiltration membranes has been empirical. We've lacked a solid framework to understand and optimize how these membranes actually work."
NF-based reactive membranes and their perceived benefits. Credit: Nature Water (2025). DOI: 10.1038/s44221-025-00467-y
Reactive nanofiltration membranes offer a powerful promise: the ability to remove salts, heavy metals and small stubborn organic pollutants all at once. But behind that promise has been a challenge—performance that's hard to predict and scale due to the complex interplay between chemical reactions and mass transport.
Reactive nanofiltration membranes offer a powerful promise: the ability to remove salts, heavy metals and small stubborn organic pollutants all at once
To tackle that complexity, Elimelech and Duan developed the first mechanistic model that simulates how oxidants and pollutants move through and react inside catalytic membranes under realistic operating conditions.
"We hypothesized that membrane performance is fundamentally governed by the interplay between reaction kinetics and solute transport," said Duan, who is now an assistant professor of civil and environmental engineering at Colorado State University. "By capturing these interactions in a model, we can move beyond trial-and-error design."
Their framework, recently published in Nature Water, simulates how variables like catalyst placement, membrane thickness, pore size and water flow affect the removal of contaminants.
"We discovered that the same membrane can behave completely differently depending on where the catalysts are located," Duan said. "At low water flux, surface-loaded catalysts dominate, but at higher flux, the action shifts inside the membrane. That has big implications for how we design systems for different water treatment needs."
One major discovery was identifying the ideal range for catalyst loading. Too little catalyst limits the reaction rate, while too much creates a transport bottleneck.
"We showed that more isn't always better," Elimelech said. "There's an optimal catalyst distribution, and we now know how to find it."
Elimelech and Duan also introduced new performance metrics that go beyond conventional removal rates—tools that can help engineers better compare and refine membrane systems across the board.
"We're shifting the field from reactive experimentation to predictive design," Duan said. "That opens the door to membranes that are not only more effective but also more scalable, energy-efficient and adaptable to different water qualities."
Importantly, the study lays out design principles for tailoring membranes to specific goals such as minimizing salt contamination, reducing energy use or maximizing contaminant selectivity. The researchers also evaluated how different oxidants, such as hydrogen peroxide and persulfate, behave inside the membrane, showing that the charge of the oxidant strongly influences its accessibility and reactivity.
"For the first time, we can simulate how changes at the molecular scale ripple out to influence full-system performance," Duan said. "That can help us build decentralized systems that serve both developed and underserved communities."
"Water is too essential to be left to guesswork," Elimelech added. "Our goal is to empower the global water community with the tools to design smarter, cleaner and more sustainable solutions."
The research was supported by the Rice Center for Membrane Excellence, the National Institutes of Health and in part by the Yale University Superfund Research Program, which is supported by a grant from the National Institute of Environmental Health Sciences.
