Commentary Evaluates New Advances in Water Transport Modeling

Northwestern Engineering’s Richard Lueptow has published a commentary article evaluating advances in computational modeling of water transport in membrane filters.

Writing in the academic journal Nature Water, Lueptow analyzes a recent paper published by researchers from Yale University, the University of Wisconsin-Madison, and Texas Tech University. That paper investigates the traditionally used solution-diffusion (SD) model, which is commonly used for modeling nanoporous reverse osmosis membranes — the dominant desalination technology.

In his commentary “Redefining a Traditional Membrane Modelling Approach,” Lueptow concluded that the work is an “immensely important contribution.” 

“After decades of depending on the SD model, researchers tend to forget that it is a greatly simplified macroscale representation of what goes on at the molecular scale: water molecules are driven by pressure through an amorphous network of molecular chains that make up the polymeric membrane barrier layer,” Lueptow wrote. “This paper confirms some existing puzzle pieces and adds several key missing ones to produce a clear and convincing picture demonstrating the weaknesses of the long-standing SD model. It further sets the conventional wisdom for continuum-based models in its proper place.”

Lueptow is senior associate dean and a professor of mechanical engineering and (by courtesy) chemical and biological engineering at the McCormick School of Engineering. His research interests and expertise range from fundamental flow physics to water purification to granular flows.

As Lueptow wrote, access to safely managed drinking water services is a pressing global need. More than 2 billion people lack access to clean water, scarcity is increasing as the population rises, climate change is wreaking more havoc, and energy demand and industrialization are continuing. These stresses mean advancing filtration technologies is even more important.

The paper “Water Transport in Reverse Osmosis Membranes Is Governed by Pore Flow, Not a Solution-Diffusion Mechanism,” published in April in Science Advances, used molecular dynamics (MD) simulations to model water flow through a reverse osmosis membrane. They found that the SD method, while useful in some cases, is flawed, showing its limitations by illustrating that water pressure linearly decreases through the thickness of the membrane, there is no water concentration gradient across the membrane, the water flux depends linearly on the applied pressure gradient, and water molecules tend to travel through the membrane in clusters. 

This work could have implications as researchers worldwide wrestle with global water insecurity.

“These models may suitably represent macroscale processes, but they are unable to capture molecular-level mechanisms. Just as important, this work demonstrates the power of MD simulations for developing the next generation of membrane technology,” Lueptow wrote. “Molecular-level simulations will be crucial for breakthroughs in membrane polymer chemistry, charge, and pore structure for optimal water transport and contaminant rejection, as well as for addressing problems of membrane scaling and biofouling, all of which are sorely needed to address the many issues associated with expanding the use and efficacy of membrane technology for water purification and treatment in our world of diminishing water resources.”