Available online at www.sciencedirect.com Journal of Membrane Science 311 (2008) 34–45 Impacts of reaction and curing conditions on polyamide composite reverse osmosis membrane properties Asim K. Ghosh, Byeong-Heon Jeong, Xiaofei Huang, Eric M.V. Hoek ∗ Civil & Environmental Engineering Department and Water Technology Research Center, University of California, Los Angeles (UCLA), Los Angeles, CA 90095-1593, USA Received 14 July 2007; received in revised form 21 November 2007; accepted 25 November 2007 Available online 4 December 2007 Abstract Here we report on the impacts of organic solvent properties, reaction conditions, and curing conditions on polyamide composite reverse osmosis membrane separation performance, film structure, and interfacial properties. We provide direct experimental evidence that: (1) MPD diffusivity in the organic phase governs MPD–TMC thin film water permeability, (2) MPD diffusivity and solubility influence MPD–TMC thin film crosslinking in competing ways, (3) water permeability correlates most strongly with MPD–TMC film structure (i.e., crosslinking), and (4) salt rejection correlates most strongly with MPD–TMC film thickness and morphology. Overall, higher flux membranes with good salt rejection appear to comprise thinner, more heavily crosslinked film structures. Such high performance RO membranes are obtained by (1) selecting high surface tension, low viscosity solvents, (2) controlling protonation of MPD and hydrolysis of TMC during interfacial polymerization, and (3) optimizing curing temperature and time based on organic solvent volatility. Finally, although more research is necessary, our results suggest the rugose morphology and relative hydrophobicity of high performance MPD–TMC membranes might enhance concentration polarization and exacerbate surface fouling. © 2007 Elsevier B.V. All rights reserved. Keywords: Reverse osmosis; Polyamide; Thin film composite; Interfacial polymerization; Desalination 1. Introduction Modern reverse osmosis (RO) membranes are formed as flat sheets or hollow fibers comprising an ultra-thin polyamide film coated over a porous polysulfone support membrane [1,2]. The selective polyamide barrier layer is formed in situ by polycon- densation reaction of polyfunctional amine and acid chloride monomers at the interface of two immiscible solvents. These elegantly engineered materials exhibit excellent performance in many desalination and water purification applications; however, significant interest remains in discovering more energy-efficient, contaminant-selective, and fouling-resistant versions of these membranes. Tailoring separation performance and interfacial properties of RO membranes requires understanding, at a fun- damental level, the mechanisms governing thin film formation. ∗ Corresponding author at: Civil & Environmental Engineering Department, 5732-G Boelter Hall, P.O. Box 951593, University of California, Los Angeles (UCLA), Los Angeles, CA 90095-1593, USA. Tel.: +1 310 206 3735; fax: +1 310 206 2222. E-mail address: hoek@seas.ucla.edu (E.M.V. Hoek). In forming a polyamide thin film, a polyfunctional amine is dissolved in water and a polyfunctional acid chloride is dissolved in apolar organic solvents like hexane, naptha, cyclohexane, freon, or isoparrafin [3–5]. When the two monomer solutions are brought into contact, both monomers partition across the liquid–liquid interface and react to form a polymer; however, polymerization occurs predominantly in the organic phase due to the relatively low solubility of most acid chlorides in water [6–8]. Therefore, it is common to use a large excess of amine over acid chloride (typically about 20:1), which drives partition- ing and diffusion of the amine into the organic phase. Any factors that alter the solubility and diffusivity of the amine monomer in the organic phase affect the reaction rate, and thus, the mor- phology and structure of the resulting polyamide film, which ultimately define separation performance and interfacial prop- erties [8,9]. Selecting the organic solvent is critical since it governs, at a minimum, the amine monomer solubility and diffusivity in the reaction zone. For example, one recent study demonstrated that hexane and isopar produced significantly different TFC mem- branes, where isopar produced RO membranes with smaller 0376-7388/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2007.11.038