Journal of Membrane Science 390–391 (2012) 211–217 Contents lists available at SciVerse ScienceDirect Journal of Membrane Science jo u rn al hom epa ge: www.elsevier.com/locate/memsci Important factors influencing molecular weight cut-off determination of membranes in organic solvents H.J. Zwijnenberg a,∗,1 , S.M. Dutczak a,b,1 , M.E. Boerrigter a , M.A. Hempenius c , M.W.J. Luiten-Olieman d , N.E. Benes b , M. Wessling b,2 , D. Stamatialis e a European Membrane Institute – Twente, MESA + Institute for Nanotechnology, University of Twente, P.O. Box 217, NL-7500 AE Enschede, The Netherlands b Membrane Science and Technology, MESA + Institute for Nanotechnology, University of Twente, P.O. Box 217, NL-7500 AE Enschede, The Netherlands c Materials Science and Technology of Polymers, MESA + Institute for Nanotechnology, University of Twente, P.O. Box 217, NL-7500 AE Enschede, The Netherlands d Inorganic Membranes, MESA + Institute for Nanotechnology, University of Twente, P.O. Box 217, NL-7500 AE Enschede, The Netherlands e Biomaterials Science & Technology, MIRA Institute for Biomedical Technology and Technical Medicine, University of Twente, P.O. Box 217, NL-7500 AE Enschede, The Netherlands a r t i c l e i n f o Article history: Received 6 June 2011 Received in revised form 27 September 2011 Accepted 21 November 2011 Available online 30 November 2011 Keywords: Solvent resistant nanofiltration MWCO characterization Flow-induced polymer deformation Polystyrene Polyisobutylene a b s t r a c t In solvent resistant nanofiltration (SRNF), sensible selection of a membrane for a particular solvent/solute system is recognized as challenging. Prospective methods for suitability analysis of membranes include molecular weight cut off (MWCO) characterization. However, insufficient understanding of the inter- related effects of solvent, solute, membrane properties, and the applied process conditions often complicates interpretation of MWCO data. This study demonstrates and discusses such effects with respect to transport mechanism. To this end very different SRNF systems have been selected: a rigid porous membrane (hydrophobized zirconia) versus a rubbery dense membrane (polydimethylsiloxane); a low flux solvent (toluene) versus high flux solvent (n-hexane); and a stiff solute (polystyrene) versus a flexible solute (polyisobutylene). The results indicate that, for the applied conditions, the MWCO of the dense membrane is predominantly affected by solute–membrane and solvent–membrane interac- tions. For the rigid porous membrane a significant effect of applied pressure is observed, in particular for the flexible solute. The non-linear relation between flux and pressure and the variations in MWCO with pressure indicate combined effects of concentration polarization and shear induced deformation of the flexible solute. The results unmistakably show that the interpretation of MWCO is heavily dependent on the system under study. © 2011 Elsevier B.V. All rights reserved. 1. Introduction In solvent resistant nanofiltration (SRNF) recently significant improvements have been achieved in the development of solvent stable membranes [1,2]. An increasing number of successful appli- cations have been reported in catalysis, the petrochemical industry and pharmaceutical industry. These applications include recovery of solvents (e.g. toluene) from dewaxed lube oil filtrates [3], sol- vent exchange [4], recovery of the organometallic complexes from various organic solvents [5], separation of phase transfer catalyst (PTC) from toluene [6,7], deacidification of vegetable oils [8,9] and concentration of pharmaceuticals [10–12]. ∗ Corresponding author. Tel.: +31 053 4893674; fax: +31 053 4894611. E-mail address: h.j.zwijnenberg@utwente.nl (H.J. Zwijnenberg). 1 Both authors contributed equally to this paper. 2 Present address: Chemische Verfahrenstechnik (CVT), RWTH Aachen University, 52064 Aachen, Germany. Amongst others, application of SRNF membranes is hampered by the complications involved in selecting appropriate membranes for each distinct separation [12]. The choice for a particular mem- brane can be based on its MWCO, however, MWCO analysis in SRNF is less developed and more complicated as compared to aqueous systems [13–20]. The complexity of MWCO in SRNF arises from pronounced effects of process conditions, inherent properties of the membrane, and solvent–solute–membrane interactions. Currently there is no universal protocol for determining the MWCO of SRNF membranes. A promising method is the filtration of a mixture of a homologues series of neutral polystyrene oligomers in toluene, as proposed by Voigt et al. [21]. This method was sys- tematically further developed by Zwijnenberg [22], See Toh et al. [23] and Dutczak et al. [24]. Based on this approach, the present work investigates the effect of process conditions and the interre- lated and inherent properties of different components. To this end, different SRNF systems have been selected: • Membrane: rigid and porous (zirconia) versus rubbery and dense (polydimethylsiloxane). 0376-7388/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2011.11.039