Conventional thin-film composite (TFC) membranes suffer from the trade-off relationship between permeability and selectivity, known as the "upper bound". In this work, we report a high performance thin-film composite membrane prepared on a tannic acid (TA)-Fe nanoscaffold (TFC) to overcome such upper bound. Specifically, a TA-Fe nanoscaffold was first coated onto a polysulfone substrate, followed by performing an interfacial polymerization reaction between trimesoyl chloride (TMC) and piperazine (PIP). The TA-Fe nanoscaffold enhanced the uptake of amine monomers and provided a platform for their controlled release. The smaller surface pore size of the TA-Fe coated substrate further eliminated the intrusion of polyamide into the substrate pores. The resulting membrane TFC showed a water permeability of 19.6 ± 0.5 L m h bar, which was an order of magnitude higher than that of control TFC membrane (2.2 ± 0.3 L m h bar). The formation of a more order polyamide rejection layer also significantly enhanced salt rejection (e.g., NaCl, MgCl, NaSO, and MgSO) and divalent to monovalent ion selectivity (e.g., NaCl/MgSO). Compared to conventional TFC nanofiltration membranes, the novel TFC membrane successfully overcame the longstanding permeability and selectivity trade-off. The current work paves a new avenue for fabricating high performance TFC membranes.

Recent studies have documented the existence of discrete voids in the thin polyamide selective layer of composite reverse osmosis membranes. Here we present compelling evidence that these nanovoids are formed by nanosized gas bubbles generated during the interfacial polymerization process. Different strategies were used to enhance or eliminate these nanobubbles in the thin polyamide film layer to tune its morphology and separation properties. Nanobubbles can endow the membrane with a foamed structure within the polyamide rejection layer that is approximately 100 nm in thickness. Simple nanofoaming methods, such as bicarbonate addition and ultrasound application, can result in a remarkable improvement in both membrane water permeability and salt rejection, thus overcoming the long-standing permeability−selectivity trade-off of desalination membranes.

Thin-film nanocomposite (TFN) membranes have been widely studied over the past decade for their desalination applications. For some cases, the incorporation of nonporous hydrophilic nanofillers has been reported to greatly enhance membrane separation performance, yet the underlying mechanism is poorly understood. The current study systematically investigates TFN membranes incorporated with silver nanoparticles (AgNPs). For the first time, we reveal the formation of nanochannels of approximately 2.5 nm in size around the AgNPs, which can be attributed to the hydrolysis of trimesoyl chloride monomers and thus the termination of interfacial polymerization by the water layer around each hydrophilic nanoparticle. These nanochannels nearly tripled the membrane water permeability for the optimal membrane. In addition, this membrane showed increased rejection against NaCl, boron, and a set of small-molecular organic compounds (e.g., propylparaben, norfloxacin, and ofloxacin), thanks to its combined effects of improved size exclusion, enhanced Donnan exclusion, and suppressed hydrophobic interaction. Our work provides fundamental insights into the formation and transport mechanisms involved in solid-filler incorporated TFN membranes. Future studies should take advantage of this spontaneous nanochannel formation in the design of TFN to overcome the classical membrane permeability–selectivity trade-off.

No abstract

Endocrine-disrupting compounds (EDCs), an important class of micropollutants with potent adverse health effects, are generally poorly rejected by traditional thin film composite polyamide membranes and thus pose significant risks in membrane-based water reclamation. We hypothesize that membrane rejection of hydrophobic EDCs can be enhanced by a hydrophilic surface coating. Using polydoamine (PDA) as a model hydrophilic coating layer, the PDA-coated NF90 membrane experienced an up to 75% reduction in the passage of bisphenol A compared to the control (NF90 without coating). Meanwhile, we also observed a systematic increase in the level of rejection of three hydrophobic parabens with an increase in PDA coating time. In contrast, there were no systematic changes in the rejection of neutral hydrophilic polyethylene glycol, which suggests that the enhanced rejection of EDCs was due to weakened EDC− membrane hydrophobic interaction. Further sorption tests revealed that the hydrophilic PDA coating could effectively decrease the rate of sorption of EDCs by the membrane, which is responsible for the improved rejection as predicted by the solution− diffusion theory. This study reveals an exciting opportunity for engineering membrane surface properties to enhance the rejection of targeted micropollutants, which has important implications in membrane-based water reclamation.

We report a fast, simple, and green coating method using the coordination complex of tannic acid (TA) and ferric ion (Fe) to enhance the removal of trace organic contaminants (TrOCs) by polyamide membranes. The entire coating process can be completed in less than 2 min; quartz crystal microbalance characterization revealed that a TA-Fe thin film formed in merely 10-20 s. Coating this TA-Fe thin film on a commercial nanofiltration membrane (NF270) reduced its effective pore size from 0.44 to 0.40 nm. The TA-Fe-coated NF270 showed significantly increased rejection of both NaCl and trace organic contaminants. In comparison with the more-time-consuming polydopamine coating (e.g., 0.5 h), the TA-Fe coating presented greater resistance to TrOC permeation (i.e., lower permeability of TrOCs). The advantages of the fast coating process, greatly improved rejection performance, and use of green accessible materials make TA-Fe a highly promising coating material for large-scale applications.

Recent studies show that the surface morphology of a thin film composite (TFC) polyamide membrane depends strongly on its porous substrate. Nevertheless, the underlining mechanisms and the effects on membrane separation performance remain controversial. To dissect the exact role of pore properties, we synthesized TFC polyamide membranes on polycarbonate substrates with cylindrical track-etched pores (PCTE) of well-defined pore size ranging from 10 to 800 nm. Leaf-like roughness features were most prominent for polyamide films formed on substrates of intermediate pore sizes (80 and 100 nm). Smaller pores inhibited leaf-like features as a result of insufficient storage of m-phenylenediamine (MPD) monomers for the interfacial reaction, whereas larger pores resulted in diminished surface roughness due to the lack of confinement to the interfacially degassed nanobubbles. Substrate porosity plays a critical role on membrane water permeability, while smaller pores with greater pore density are favored to improve membrane rejection. TFC polyamide membranes prepared on sponge-like poly­(ether sulfone) and polysulfone substrates exhibit better water permeability and salt rejection compared to the PCTE-TFC membranes thanks to the simultaneously enhanced confinement and MPD storage effects. The mechanistic insights gained in this study reveal the huge potential of substrate design toward high-performance TFC RO membranes.

Polyamide-based thin film composite (TFC) membranes are generally optimized for salt rejection but not for the removal of trace organic contaminants (TrOCs). The insufficient rejection of TrOCs such as endocrine disrupting compounds (EDCs) by polyamide membranes can jeopardize product water safety in wastewater reclamation. In this study, we report a novel nonpolyamide membrane chemistry using green tannic acid−iron (TA−Fe) complexes to remove TrOCs. The nanofiltration membrane formed at a TA−Fe molar ratio of 1:3 (TA−Fe3) had a continuous thin rejection layer of 10−30 nm in thickness, together with a water permeability of 5.1 Lm 2− h −1 bar −1 and a Na 2 SO 4 rejection of 89.7%. Meanwhile, this membrane presented significantly higher rejection of EDCs (up to 99.7%) than that of polyamide membranes (up to 81.8%). Quartz crystal microbalance results revealed that the sorption amount of a model EDC, benzylparbaen, by TA−Fe3 layer was nearly 2 orders of magnitude less than that by polyamide, leading to reduced transmission and higher rejection. Further analysis of membrane revealed a much greater water/EDC selectivity of the TA−Fe3 membrane compared to the polyamide membranes.