Figure 1. This is a tree-like system of hexagonal nanopores showing channel flow. Courtesy of Flickr.
When it comes to any filter—a sieve, a water filter, or even a tissue membrane—your instinct probably tells you that the tighter the pores, the better the filtration. However, a recent study from the Yale School of Engineering & Applied Sciences suggests otherwise. Researchers Brian Shoemaker, Omar Khalifa, and Amir Haji-Akbari revealed that pore density, a measure of how tightly the pores are packed together in a filter, can actually be detrimental to effective filtration.
“In early 2020, a new student [Brian Shoemaker] and I suggested putting two pores next to each other to investigate our methods, expecting them to double the flux of a single pore system,” Amir said. “Yet our observations contradicted this.” They sought to revolutionize gas separation and water filtration in membranes by optimizing the geometry of their pores to achieve high permeability and selectivity.
They conducted simulations of the complex flow of molecules across various filter types and configurations, or arrangements of pores. The results showed that closely-spaced pores increase co-ion transport but reduce counter-ion transport, meaning that very densely packed nanopores allow more unwanted particles through than expected.
This research may help build better filtration systems for desalination plants, to make fresh water more accessible, and for polluting factories, to trap toxic chemicals in gases before they are released into the atmosphere. Furthermore, the model could be used to design membranes for capsules that release drugs directly into organs under certain conditions. These findings pave the way towards smarter filtration solutions that are as effective in application as they are in theory.
[Citations]
Shoemaker, B. A., Khalifa, O., & Haji-Akbari, A. (2024). Correlations in charged multipore systems: Implications for enhancing selectivity and permeability in nanoporous membranes. ACS Nano, 18(2), 1420-1431. https://doi.org/10.1021/acsnano.3c07489