Intercrystalline Channels at Subnanometer Scale for Precise Molecular Nanofiltration.

Membrane-based technologies can provide cost-effective and energy-efficient methods for various separation processes. The key goal is to develop materials with uniform, tunable, and well-defined subnanometer-scale channels. Suitable membrane materials should have high selectivity and permeance and can be manufactured in a robust and scalable fashion. Here, we report the construction of sub-1 nm intercrystalline channels with such characteristics and elucidate their transport properties. These channels are formed by assembling 3D aluminum formate crystals during the amorphous-to-crystalline transformation process. By controlling the transformation time, the channel size can be tuned from the macroscopic scale to nanometer scale. The resulting membranes exhibit tailored selectivity and permeance, with molecular weight cutoffs ranging from around 300 Da to approximately 650 Da, and ethanol permeance ranging from 0.8 to 22.0 L m-2 h-1 bar-1. We further show that liquid flow through these channels changes from viscosity-dominated continuum flow to subcontinuum flow, which can be described by a modified Hagen-Poiseuille model. Our strategy provides a new scalable platform for applications that commonly exploit nanoscale mass transport.

Polyamide membranes with nanoscale ordered structures for fast permeation and highly selective ion-ion separation.

Fast permeation and effective solute-solute separation provide the opportunities for sustainable water treatment, but they are hindered by ineffective membranes. We present here the construction of a nanofiltration membrane with fast permeation, high rejection, and precise Cl-/SO42- separation by spatial and temporal control of interfacial polymerization via graphitic carbon nitride (g-C3N4). The g-C3N4 nanosheet binds preferentially with piperazine and tiles the water-hexane interface as revealed by molecular dynamics studies, thus lowering the diffusion rate of PIP by one order of magnitude and restricting its diffusion pathways towards the hexane phase. As a result, membranes with nanoscale ordered hollow structure are created. Transport mechanism across the structure is clarified using computational fluid dynamics simulation. Increased surface area, lower thickness, and a hollow ordered structure are identified as the key contributors to the water permeance of 105 L m2·h-1·bar-1 with a Na2SO4 rejection of 99.4% and a Cl-/SO42- selectivity of 130, which is superior to state-of-the-art NF membranes. Our approach for tuning the membrane microstructure enables the development of ultra-permeability and excellent selectivity for ion-ion separation, water purification, desalination, and organics removal.

Macropatterning of Microcrumpled Nanofiltration Membranes by Spacer Imprinting for Low-Scaling Desalination.

Surface patterns provide a chemical-free approach to reduce fouling by mimicking nature and are yet limited by their complicated fabrication procedures. Here, we report readily scalable methods to create sub-micrometer- and millimeter-scale patterns on membrane surfaces for low-scaling desalination, with a focus on the antiscaling mechanism. Specifically, a robust polyethylene (PE) lithium battery separator prepared from melt casting and stretching has been used as the support for nanofiltration (NF), giving micrometer-scale crumples on the surface. Then, the PENF membrane is imprinted by a permeate spacer during tests, leading to millimeter-scale patterns. Two types of experiments are designed to give insights into the impact of surface structure on scaling in NF processes, including (1) comparisons of smooth surfaces and surfaces with nanometer-, micrometer-, and millimeter-scale features and (2) no-stirring dead-end tests and crossflow tests. It has been found that micrometer-scale patterns are resistant to scaling through both spatial and hydrodynamic effects, and millimeter-scale patterns are also effective in reducing scaling solely due to hydrodynamic effects. Computational fluid dynamics (CFD) simulation gives further explanations. In addition, organic and microbial fouling has been studied to give implications for future membrane engineering.

Understanding the Roughness-Fouling Relationship in Reverse Osmosis: Mechanism and Implications.

This paper investigates the relationship between roughness and fouling in reverse osmosis (RO) through specially designed experimental protocols and computational fluid dynamics (CFD) studies. Conventional polyamide thin-film composite membranes with a ridge-and-valley structure and an emerging type of smooth membranes are prepared. A wide range of foulants are used, and fouling under static, crossflow, and RO conditions are tested. Feature size on the membrane surface is important when foulants and the microscale structure show a similar size, and otherwise membrane-foulant interactions govern the static attachment. Under crossflow mode, fouling on the ridge-and-valley surface is not reduced to the same extent as that on smooth membranes, with the insufficient vortices in the valley region being identified as the key factor by CFD studies. In RO, uneven flux distribution as confirmed by gold nanoparticle filtration is also found to account for the much higher fouling rate of conventional membranes. Our study then suggests two strategies to design next-generation fouling-resistant RO membranes via structural optimization: first, a smooth selective layer should be maintained to ensure uniform flux distribution; second, one may mimic nature to fabricate patterned porous membranes as the support, so that it optimizes hydrodynamics while maintaining even fluxes.