Reverse osmosis (RO) is currently the most important desalination technology and it is experiencing significant growth. The objective of this paper is to review the historical and current development of RO membrane materials which are the key determinant of separation performance and water productivity, and to define the targets for those who are developing new RO membrane materials. The chemistry, synthesis mechanism(s) and desalination performance of various RO membranes are discussed from the point of view of membrane materials science. The review starts with the first generation of asymmetric polymeric membranes and finishes with current proposals for nano-structured membrane materials. The review provides an overview of RO performance in relation to membrane materials and methods of synthesis. To date polymeric membranes have dominated the RO desalination industry. From the late 1950s to the 1980s the research effort focussed on the search for optimum polymeric membrane materials. In subsequent decades the performance of RO membranes has been optimised via control of membrane formation reactions, and the use of poly-condensation catalysts and additives. The performance of the state-of-the-art RO membrane has been highlighted. Nevertheless, the advances in membrane permselectivity in the past decade has been relatively slow, and membrane fouling remains a severe problem. The emergence of nano-technology in membrane materials science could offer an attractive alternative to polymeric materials. Hence nano-structured membranes are discussed in this review including zeolite membranes, thin film nano-composite membranes, carbon nano-tube membranes, and biomimetic membranes. It is proposed that these novel materials represent the most likely opportunities for enhanced RO desalination performance in the future, but that a number of challenges remain with regard to their practical implementation.

All published reports on fluid flow enhancement and water slippage are associated with hydrophobic surfaces, such as carbon nanotubes. Here, we investigate water flow in hydrophilic alumina nanochannels with diameters ranging from 20 nm to 100 nm. For the smallest channels tested, the water permeability is more than double than the theoretical prediction using the Hagen-Poiseuille equation. Though such an enhancement is significantly smaller than what has been measured in carbon nanotubes, it clearly shows that flow enhancement and water slippage occurs on hydrophilic surfaces as well, contrary to existing theoretical models. To the authors' knowledge, it is the first experimental demonstration of water slippage on hydrophilic surfaces. The results show the dependence of the flow enhancement on the surface chemistry, diameter and length of the nanochannel.

No abstract

Since their discovery, carbon nanotubes have been considered as a potential material for filtration applications due to low tortuosity, smooth structure and the possibility of fine tuning their diameter. Measurements of fluid flow in nanotubes, with diameters ranging from 0.6 to 100 nm dramatically raised interest in them, with very high water flow rates promising to deliver orders-ofmagnitude higher performance compared to other membranes. This promise was based on reports of flow enhancement, defined as a ratio of the measured flow compared to a no-slip Poiseuille flow, ranging from 10 to 100,000 with the underlying assumption that commercial membranes would exhibit the no-slip behavior. The concept of flow enhancement, though, is of little help for actual filtration applications where one is interested in a membrane's performance in terms of selectivity and permeability. In this work, the flow enhancement and permeability of UF carbon nanotube-anodic alumina membranes (CNT-AAM) with a large range of diameters is reported. Using a recently developed model, it is shown that the permeability is directly related to the solidliquid molecular interactions between the liquid and the nanotubes. Finally, the performance of these CNT membranes and others in the literature has been analyzed in terms of permeability, comparing them to commercial membranes in the RO, NF and UF ranges. Results show that in fact, carbon nanotube membranes have a higher pure water permeability than commercial polymer membranes.

No abstract

This paper investigates the effect of surface structure and chemistry on the wetting properties of nanostructured porous anodic alumina (PAA). Measurements of the equilibrium apparent contact angle (APCA) were first taken on as produced hydrophilic nanoporous alumina with a range of pore diameters from 10 to 170 nm, yielding a range of contact angles from 10 to 100°. The PAAs were then coated with a fluorosilane to change the surface chemistry of the nanostructures. The same trend was observed as in the hydrophilic case, but the contact angles increased from 106 to 150° for pores sizes ranging from 10 to 100 nm for the hydrophobic PAA. These results probe the limits of the current wetting models such as the Cassie-Baxter and Wenzel equations for nanostructured materials. A geometric model has been developed using the equation proposed by Marmur to explain the wetting properties of the bare-and silanized-PAA.

No abstract

Tubular alumina membranes exhibiting symmetric and asymmetric morphology were fabricated via electrochemical anodization of aluminium and studied for ultrafiltration application. By controlling the anodization conditions, the pore structure can be precisely controlled at the nanometre scale. Via reduction of the anodization voltage in a sudden or gradual manner, two types of asymmetric cross-section morphologies were obtained. The membranes were characterized by MWCO, pure water permeability, bovine serum albumin (BSA) rejection and fouling tests. The breakdown of linear relationship between anodization voltage and pore diameter was observed for anodization below 10 V. The selectivitypermeability analysis was compared to the framework developed by Metha and Zydney (JMS, 2005). The analysis shows the asymmetric membranes still suffering from low permeability despite providing good rejection properties. Most of the resistance to water permeability is, however, contributed by the thickness of the support layer of the membranes. The flux decline during BSA filtration can be modelled using combined complete pore blocking -cake filtration model. Although the membranes show good rejection performance and potential for scale-up application by fabrication in tubular form, further improvement in permeability and fouling mitigation could be achieved by developing a more porous support layer and surface modification. Draft-Please refer to published article only for details and data 3 Research Highlights  Fabrication of tubular asymmetric anodized alumina membranes for ultrafiltration applications  The membranes have uniform pore structure which is readily controllable  The membranes exhibit good bovine serum albumin rejection Draft -Please refer to published article only for details and data