Journal of Membrane Science 372 (2011) 366–372 Contents lists available at ScienceDirect Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci Liquid–liquid displacement porosimetry for the characterization of virus retentive membranes René Israel Peinador a , José Ignacio Calvo a,∗ , Khuong ToVinh b , Volkmar Thom b , Pedro Prádanos a , Antonio Hernández a a SMAP (UA-UVA-CSIC), Departamento de Física Aplicada, Facultad Ciencias, Universidad de Valladolid, Real de Burgos, s/n, 47071 Valladolid, Spain b Sartorius-Stedim Biotech GmbH, Membrane R&D Biotechnology, August Spindler-Str. 11, 37079 Göttingen, Germany article info Article history: Received 17 September 2010 Received in revised form 14 February 2011 Accepted 16 February 2011 Available online 22 February 2011 Keywords: Virus retentive membrane Dextran retention Phage retention Pore size distribution Liquid–Liquid porosimetry abstract Parvovirus retentive membranes made from polyethersulfone (PES) have been characterized by differ- ent techniques including dextran and phage retention. Results have been correlated with the pore size distributions of such membranes as obtained by liquid–liquid displacement porosimetry (LLDP). The results of LLDP concerning pore size distributions are proved to be consistent with those obtained by image analysis of SEM transversal sections and refer to the narrower pore section. Moreover, the maximum pore size determined by LLDP fairly correlates with the measured retention capabilities of the membranes. LLDP results suggest that the technique can be an accurate method for the determination of pore size characteristics of virus retentive membranes. This technique can be simplified to be even faster and straightforward by detecting only the maximum pore size of the membrane. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Mammalian cell cultures used in the production of monoclonal antibodies and therapeutic recombinant proteins are vulnerable to contamination by viruses. Plasma-derived pharmaceuticals can also be potentially infected by viral pathogens [1]. Due to these vulnerabilities, biopharmaceutical manufacturing procedures need robust and efficient purification steps to prevent microbiologi- cal contamination of the products [2]. Chlorination has been in the past a very popular technique for quality control of treated water in terms of viral/microbiological safety. Nevertheless, chlo- rination can give rise to undesired by-products and an adequate control of dosage is difficult, especially in small scale plants [3]. Finally, most viruses are more resistant to chlorine than bacteria. Another possibility relies in size exclusion filtration using virus retentive membrane filters that are used mainly in downstream- processing of pharmaceutical solutions. Virus retentive membranes have severe requirements, as they must remove more than 99.9% of virus particles while passing almost the entire protein product in the feed stream [4]. The difference in size between a parvovirus and an antibody is relatively small, making size-based virus clearance a challenging technology. ∗ Corresponding author. Tel.: +34 983423758; fax: +34 983423136. E-mail address: jicalvo@termo.uva.es (J.I. Calvo). Several degrees of virus removal have been reported for UF and even MF membranes [5–8], or membrane adsorbers [9]. Viruses could be expected to be completely retained by tight UF mem- branes (with molecular weight cut-offs of 10–100 kDa), according to the molecular weight of the viruses [10]. Nevertheless, it has been reported, that small viruses have penetrated membranes theoreti- cally included in the NF range [3], while log reduction values (see Eq. (1) below) in the 6–7 range can be obtained with pre-treatment through MF membranes [11,12]. The manufacturers of virus retentive membranes classify the virus clearance filters into two broad categories based on the removal needs of the biotechnological industry – filters that are capable of removing 50 nm or larger viruses (retroviruses) and fil- ters that can remove both small (∼20 nm parvoviruses) and large viruses. Virus retentive membranes typically exhibit a pore size gradi- ent, where the pore size increases from the skin layer progressing towards a large pore microfiltration layer. The skin provides the selectivity needed to exclude viruses, while the thicker support layer provides mechanical support for the membrane. Obviously, the advantage of a thin retentive membrane skin consists in a large overall flux. The overall LRV of a membrane could be increased simply using multiple layered membrane with the subsequent reduction in flux or the use of membranes with inner narrowing of the pores which is effectively the approach used in the membranes studied here. Others studies focus on the evaluation of affinity-type membranes which show a case-by-case virus removal ability [13]. 0376-7388/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2011.02.022