An international team led by researchers at KU Leuven, with contributing authors from the University of Bath, Nanjing University, Huazhong University of Science and Technology and Monash University, has developed a graphene oxide membrane that speeds up the removal of water from isopropanol (IPA), a solvent used worldwide in the pharmaceutical and electronics industries. The results offer an alternative to conventional distillation-based purification, which requires high energy input.
Structural design of N-GOm: GO (gray) and NPGO (blue) nanosheets on a nylon substrate (yellow), forming sp2 graphitic and hydrophilic sp3 domains that create the membrane's cavity structure. Image credit: Nature Communications
Separating chemical mixtures into pure components is a crucial but energy-intensive process in industrial chemistry, accounting for 10 to 15% of global energy use. IPA, produced globally at over 3.5 million metric tons a year in a market exceeding $6.3 billion, is typically purified from water-laden mixtures using heating and distillation, methods that carry a significant energy and CO2 footprint. The team turned instead to pervaporation, a membrane-based separation technique that bypasses vapor-liquid equilibrium constraints and uses only the latent heat of evaporation. The membrane, termed N-GOm, is built by co-assembling conventional graphene oxide (GO) nanosheets with a newly developed variant, nanoporous graphene oxide (NPGO), whose sheets carry smaller pores and oxygen-rich functional groups that increase water affinity. Combining the two nanosheet types creates an internal structure with two functions: narrow channels that block larger molecules, and hydrophilic regions that attract and transport water. The researchers report that this design raises the DFT-calculated water adsorption energy roughly 2.6-fold and lowers the diffusion energy barrier by about 40% compared with standard GO membranes.
After thermal crosslinking, the resulting membrane (rN-GOm) achieved a water flux of 18.4 kg·m⁻²·h⁻¹ when treating a 70/30 wt% IPA/water feed, and produced a permeate with roughly 99.6 wt% water purity from a 90/10 wt% IPA/water feed, a flux the authors say is 3 to 10 times higher than conventional pervaporation membranes reported in the literature. Long-term testing showed stable separation performance with minimal decay.
"The main challenge is to design a structure with channels that are not too small, which would slow down the separation and require more energy, but also not too large, which would reduce the purity of the final product," said Lei Jiang, a doctoral researcher at KU Leuven. "The new membrane combines both efficient and high-quality separation in a single structure."
"The new membrane efficiently removes water from a mixture containing 90% isopropanol and 10% water. It selectively transports water through the membrane, leaving the isopropanol-rich mixture increasingly dehydrated and therefore more suitable for recovery and reuse," said Dr. Pengrui Jin, a Prize Fellow in the Department of Chemical Engineering at the University of Bath and an independent principal investigator on the study. "The separated permeate contains about 99.6% water, demonstrating the membrane's high selectivity for water. In addition, the process is faster than existing techniques and requires less energy, as it does not rely on high temperatures."
Professor Bart Van der Bruggen of KU Leuven, who led the project, said the technology could support the shift toward more sustainable solvent production, including bio-based IPA made from renewable feedstocks. "The membrane delivers gains across the board: purity, energy consumption and economic efficiency," he said. "We are eager to test the technology on other chemical mixtures as well."
The team is now exploring options to scale up the membrane and is assessing the possibility of filing a patent.
