A randomly distributed single-walled carbon nanotube reverse osmosis membrane for seawater desalination: microstructural design

Abstract

Compared with arrayed carbon nanotubes (CNTs), randomly arranged CNTs are much more practical (only 1/10 the price of arrayed CNTs). This study employs molecular dynamics (MD) simulations to investigate the desalination performance of pure random carbon nanotube (CNT) membranes with varying inner diameters ranging from 0.8 nm to 1.3 nm. Modeling and micro-separation processes show that the random distribution makes the CNT channels complex and diverse, and water molecules have multiple paths (inside the tube, outside the tube, and a combination of inside and outside the tube) when passing through membranes. It was found that the water flux of random (7,7) CNT membranes reached six times that of the arrayed CNTs. For ions, the path through the membrane becomes intricate and elongated, akin to traversing a maze. The randomized (10,10) CNT membrane retains ions six times more effectively than the arrayed CNTs. Remarkably, it was also observed that the randomized (8,8) CNT membranes mitigated the trade-off effect to some extent. Stacking two sizes of CNTs enhances the pore size characteristics. It was observed that 80% of the composite system can achieve a water flux of 50 L cm⁻² day⁻¹ MPa⁻¹, while 40% of the composite system exhibits a retention rate exceeding 80%. The results show that the localized pore size superposition effect is highly desirable, which improves the water flux and retention rate to different degrees and achieves the effect of 1 + 1 > 2, and a novel reference is presented for the application of CNTs as a functional layer in seawater desalination processes.