At Tohoku University, researchers have cracked a problem that has vexed materials scientists for over a decade: they've engineered membranes that separate carbon dioxide more efficiently than anyone thought possible, breaking through a performance ceiling that's stood since 2008.
The challenge sounds simple but has long been stubbornly hard. When you need to filter carbon dioxide from natural gas, purify hydrogen, or capture carbon from the air, you face an impossible choice. Make a membrane that lets CO₂ flow through quickly, and it loses its ability to distinguish CO₂ from other gases like methane and hydrogen. Make it selective enough to separate effectively, and the CO₂ moves through like molasses. This bind—the permeability-selectivity trade-off—has been the inescapable law of membrane chemistry.
Enter covalent organic frameworks, or COFs. These are crystalline porous materials with atomic-scale precision, built so their pores can be tuned to attract specific molecules. The Tohoku team, led by Dr. Saikat Das and Yuichi Negishi at the Institute of Multidisciplinary Research for Advanced Materials, took an elegant approach: they designed two nearly identical porous materials that differed in just one detail—the type of atoms lining their pores. One, called TUS-621, was engineered with oxygen-rich surfaces. The other, TUS-622, used sulfur instead. This allowed them to isolate exactly how chemistry, not structure, shapes gas separation.
The oxygen-rich TUS-621 proved transformative. When embedded into a polymer membrane in a composite known as a mixed matrix membrane, or MMM, it created pathways that both attracted CO₂ molecules and allowed them to speed through. The best-performing version—TUS-621 combined with a polymer called Pebax at 10 percent loading—did something remarkable: it surpassed the 2008 Robeson upper bound, a benchmark that had seemed like the inherent ceiling for membrane performance. More impressive still, the membrane maintained this exceptional separation performance continuously for 30 days while operating across broad ranges of pressure and temperature, and it excelled at separating CO₂ from both methane and hydrogen.
The breakthrough matters because current industrial methods for CO₂ separation—amine scrubbing and cryogenic processes—are energy hogs. They demand significant heat and cooling, making them expensive and carbon-intensive, exactly what you don't want when trying to reduce emissions. Membrane-based alternatives could transform industries from natural gas processing to hydrogen production to direct air capture, all while using far less energy.
Computational analysis revealed why oxygen worked so much better than sulfur: the oxygen-rich pore surfaces create stronger electronic coupling with CO₂ molecules, essentially giving them a preferential pathway through the membrane while slowing down competing gases. It's chemistry at the atomic level reshaping behavior at the macroscopic scale.
"This study demonstrates that precise heteroatom engineering within structurally controlled COFs can fundamentally reshape membrane transport behavior," Negishi said in describing the work, published in the Journal of the American Chemical Society. The team believes this strategy opens a genuine pathway toward practical, energy-efficient carbon capture and gas separation. After decades of bumping against the same ceiling, researchers have finally found the door.